記述言語：日本語   出版者・発行元：(公社)日本生物工学会  

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：MDPI AG  

Bayesian optimization (BO)-assisted screening was applied to identify improved reaction conditions toward a hundred-gram scale-up synthesis of 2,3,7,8-tetrathiaspiro[4.4]nonane (1), a key synthetic intermediate of 2,2-bis(mercaptomethyl)propane-1,3-dithiol [tetramercaptan pentaerythritol]. Starting from the initial training set (ITS) consisting of six trials sampled by random screening for BO, suitable parameters were predicted (78% conversion yield of spiro-dithiolane 1) within seven experiments. Moreover, BO-assisted screening with the ITS selected by Latin hypercube sampling (LHS) further improved the yield of 1 to 89% within the eight trials. The established conditions were confirmed to be satisfactory for a hundred grams scale-up synthesis of 1.

DOI： 10.3390/molecules28135180

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Science and Business Media LLC  

Abstract

We aimed to identify combinations of clinical factors that predict heart failure (HF) onset using a novel limitless-arity multiple-testing procedure (LAMP). We also determined if increases in numbers of predictive combinations of factors increases the probability of developing HF. We recruited people without HF who received health check-ups in 2010, who were followed annually for 4 years. Using 32,547 people, LAMP was performed to identify combinations of factors of fewer than four factors that could predict the onset of HF. The ability of the method to predict the probability of HF onset based on the number of matching predictive combinations of factors was determined in 275,658 people. We identified 549 combinations of factors for the onset of HF. Then we classified 275,658 people into six groups who had 0, 1–50, 51–100, 101–150, 151–200 or 201–250 predictive combinations of factors for the onset of HF. We found that the probability of HF progressively increased as the number of predictive combinations of factors increased. We identified combinations of variables that predict HF onset. An increased number of matching predictive combinations for the onset of HF increased the probability of HF onset.

DOI： 10.1038/s41598-023-31600-0

researchmap

その他リンク： https://www.nature.com/articles/s41598-023-31600-0

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1021/acs.oprd.2c00267

researchmap

掲載種別：研究論文（学術雑誌）  

Traditional optimization methods using one variable at a time approach waste time and chemicals and assume that different parameters are independent from one another. Hence, a simpler, more practical, and rapid process for predicting reaction conditions that can be applied to several manufacturing environmentally sustainable processes is highly desirable. In this study, biaryl compounds were synthesized efficiently using an organic Brønsted acid catalyst in a flow system. Bayesian optimization-assisted multi-parameter screening, which employs one-hot encoding and appropriate acquisition function, rapidly predicted the suitable conditions for the synthesis of 2-amino-2′-hydroxy-biaryls (maximum yield of 96%). The established protocol was also applied in an optimization process for the efficient synthesis of 2,2′-dihydroxy biaryls (up to 97% yield). The optimized reaction conditions were successfully applied to gram-scale synthesis. We believe our algorithm can be beneficial as it can screen a reactor design without complicated quantification and descriptors.

DOI： 10.1038/s42004-022-00764-7

Scopus

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Science and Business Media LLC  

DOI： 10.1007/s10115-022-01765-7

researchmap

その他リンク： https://link.springer.com/article/10.1007/s10115-022-01765-7/fulltext.html

掲載種別：研究論文（学術雑誌）   出版者・発行元：Frontiers Media SA  

The emergence of bacteria that are resistant to antibiotics is common in areas where antibiotics are used widely. The current standard procedure for detecting bacterial drug resistance is based on bacterial growth under antibiotic treatments. Here we describe the morphological changes in enoxacin-resistant Escherichia coli cells and the computational method used to identify these resistant cells in transmission electron microscopy (TEM) images without using antibiotics. Our approach was to create patches from TEM images of enoxacin-sensitive and enoxacin-resistant E. coli strains, use a convolutional neural network for patch classification, and identify the strains on the basis of the classification results. The proposed method was highly accurate in classifying cells, achieving an accuracy rate of 0.94. Using a gradient-weighted class activation mapping to visualize the region of interest, enoxacin-resistant and enoxacin-sensitive cells were characterized by comparing differences in the envelope. Moreover, Pearson’s correlation coefficients suggested that four genes, including lpp, the gene encoding the major outer membrane lipoprotein, were strongly associated with the image features of enoxacin-resistant cells.

DOI： 10.3389/fmicb.2022.839718

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Elsevier BV  

DOI： 10.1016/j.commatsci.2022.111207

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：ROYAL SOC CHEMISTRY  

Multiparameter screening of reductive carboxylation in an electrochemical flow microreactor was performed using a Bayesian optimization (BO) strategy. The developed algorithm features a constraint on passed charge for the electrochemical reaction, which led to suitable conditions being instantaneously found for the desired reaction. Analysis of the BO-suggested conditions underscored the physicochemical validity.

DOI： 10.1039/d2cc00124a

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Science and Business Media LLC  

<title>Abstract</title>High-throughput, high-accuracy detection of emerging viruses allows for the control of disease outbreaks. Currently, reverse transcription-polymerase chain reaction (RT-PCR) is currently the most-widely used technology to diagnose the presence of SARS-CoV-2. However, RT-PCR requires the extraction of viral RNA from clinical specimens to obtain high sensitivity. Here, we report a method for detecting novel coronaviruses with high sensitivity by using nanopores together with artificial intelligence, a relatively simple procedure that does not require RNA extraction. Our final platform, which we call the artificially intelligent nanopore, consists of machine learning software on a server, a portable high-speed and high-precision current measuring instrument, and scalable, cost-effective semiconducting nanopore modules. We show that artificially intelligent nanopores are successful in accurately identifying four types of coronaviruses similar in size, HCoV-229E, SARS-CoV, MERS-CoV, and SARS-CoV-2. Detection of SARS-CoV-2 in saliva specimen is achieved with a sensitivity of 90% and specificity of 96% with a 5-minute measurement.

DOI： 10.1038/s41467-021-24001-2

researchmap

その他リンク： http://www.nature.com/articles/s41467-021-24001-2

掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Science and Business Media LLC  

<title>Abstract</title>Controlling the fast electrophoresis of nano-objects in solid-state nanopores is a critical issue for achieving electrical analysis of single-particles by ionic current. In particular, it is crucial to slow-down the translocation dynamics of nanoparticles. We herein report that a focused electric field and associated water flow in a surround-gate nanopore can be used to trap and manipulate a nanoscale object. We fine-control the electroosmosis-induced water flow by modulating the wall surface potential via gate voltage. We find that a nanoparticle can be captured in the vicinity of the conduit by balancing the counteracting electrophoretic and hydrodynamic drag forces. By creating a subtle force imbalance, in addition, we also demonstrate a gate-controllable motion of single-particles moving at an extremely slow speed of several tens of nanometers per second. The present method may be useful in single-molecule detection by solid-state nanopores and nanochannels.

DOI： 10.1038/s43246-021-00132-3

researchmap

その他リンク： http://www.nature.com/articles/s43246-021-00132-3

掲載種別：研究論文（学術雑誌）   出版者・発行元：American Chemical Society (ACS)  

DOI： 10.1021/acs.joc.1c01242

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Wiley  

DOI： 10.1002/smtd.202100542

researchmap

その他リンク： https://onlinelibrary.wiley.com/doi/full-xml/10.1002/smtd.202100542

掲載種別：研究論文（学術雑誌）   出版者・発行元：MDPI AG  

A better understanding of the microstructure–property relationship can be achieved by sampling and analyzing a microstructure leading to a desired material property. During the simulation of filled rubber, this approach includes extracting common aggregates from a complex filler morphology consisting of hundreds of filler particles. However, a method for extracting a core structure that determines the rubber mechanical properties has not been established yet. In this study, we analyzed complex filler morphologies that generated extremely high stress using two machine learning techniques. First, filler morphology was quantified by persistent homology and then vectorized using persistence image as the input data. After that, a binary classification model involving logistic regression analysis was developed by training a dataset consisting of the vectorized morphology and stress-based class. The filler aggregates contributing to the desired mechanical properties were extracted based on the trained regression coefficients. Second, a convolutional neural network was employed to establish a classification model by training a dataset containing the imaged filler morphology and class. The aggregates strongly contributing to stress generation were extracted by a kernel. The aggregates extracted by both models were compared, and their shapes and distributions producing high stress levels were discussed. Finally, we confirmed the effects of the extracted aggregates on the mechanical property, namely the validity of the proposed method for extracting stress-contributing fillers, by performing coarse-grained molecular dynamics simulations.

DOI： 10.3390/polym13162683

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Wiley  

DOI： 10.1002/smtd.202100191

researchmap

その他リンク： https://onlinelibrary.wiley.com/doi/full-xml/10.1002/smtd.202100191

掲載種別：研究論文（学術雑誌）   出版者・発行元：IOS Press  

We propose novel approaches for classification from positive and unlabeled data (PUC) based on maximum likelihood principle. These are particularly suited to measurement tasks in which the class prior of the target object in each measurement is unknown and significantly different from the class prior used for training, while the likelihood function representing the observation process is invariant over the training and measurement stages. Our PUCs effectively work without estimating the class priors of the unlabeled objects. First, we present a PUC approach called Naive Likelihood PUC (NL-PUC) using the maximum likelihood principle in a nontrivial but rather straightforward manner. The extended version called Enhanced Likelihood PUC (EL-PUC) employs an algorithm iteratively improving the likelihood estimation of the positive class. This is advantageous when the availability of the labeled positive data is limited. These characteristics are demonstrated both theoretically and experimentally. Moreover, the practicality of our PUCs is demonstrated in a real application to single molecule measurement.

DOI： 10.3233/ida-194980

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Reversibly photoswitchable fluorescent proteins (RSFPs) are a class of fluorescent proteins whose fluorescence can be turned on and off by light irradiation. RSFPs have become essential tools for super-resolution (SR) imaging. Because most SR imaging techniques require high power density illumination, mitigating phototoxicity in cells due to intense light irradiation has been a challenge. Although we previously developed a RSFP named Kohinoor to achieve SR imaging with low phototoxicity, the photoproperties were insufficient to move a step further to explore the cellular dynamics by SR imaging. Here, we show an improved version of RSFP, Kohinoor2.0, which is suitable for SR imaging of cellular processes. Kohinoor2.0 shows a 2.6-fold higher fluorescence intensity, 2.5-fold faster chromophore maturation, and 1.5-fold faster off-switching than Kohinoor. The analysis of the pH-dependence of the visible absorption band revealed that Kohinoor2.0 and Kohinoor were in equilibria among multiple fluorescently-bright and dark states, with the mutations introduced into Kohinoor2.0 bringing about a higher stabilization of the fluorescently-bright states compared to Kohinoor. Using Kohinoor2.0 with our SR imaging technique, SPoD-OnSPAN, we conducted 4-h time-lapse SR imaging of an actin filament network in mammalian cells with a total acquisition time of 480 s without a noticeable indication of phototoxicity. Furthermore, we demonstrated the SR imaging of mitochondria dynamics at a time resolution of 0.5 s, in which the fusion and fission processes were clearly visualized. Thus, Kohinoor2.0 is shown to be an invaluable RSFP for the SR imaging of cellular dynamics.

DOI： 10.1093/jmicro/dfab001

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC  

In this study, we developed an odor sensor system using chemosensitive resistors, which outputted multichannel data. Mixtures of gas chromatography stationary materials (GC materials) and carbon black were used as the chemosensitive resistors. The interaction between the chemosensitive resistors and gas species shifted the electrical resistance of the resistors. Sixteen different chemosensitive resistors were fabricated on an odor sensor chip. In addition, a compact measurement instrument was fabricated. Sixteen channel data were obtained from the measurements of gas species using the instrument. The data were analyzed using machine learning algorithms available on Weka software. As a result, the sensor system successfully identified alcoholic beverages. Finally, we demonstrated the classification of restroom odor in a field test. The classification was successful with an accuracy of 97.9%.

DOI： 10.1109/JSEN.2020.3016678

Web of Science

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1021/acs.analchem.0c04353

PubMed

researchmap

担当区分：最終著者   掲載種別：論文集(書籍)内論文   出版者・発行元：Springer International Publishing  

DOI： 10.1007/978-3-030-75015-2_5

researchmap

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr2109.html#abs-2109-14198

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

DOI： 10.1109/ICDM51629.2021.00073

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2021.html#TingWWZ21

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10618-021-00785-1

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Royal Society of Chemistry (RSC)  

A highly efficient synthesis of α-ketiminophosphonates has been established for the electrochemical oxidation of α-amino phosphonates with the utilization of machine-learning-assisted simultaneous multiparameter screening.

DOI： 10.1039/d1gc01583d

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Science and Business Media LLC  

<title>Abstract</title>
A rapid method for screening pathogens can revolutionize health care by enabling infection control through medication before symptom. Here we report on label-free single-cell identifications of clinically-important pathogenic bacteria by using a polymer-integrated low thickness-to-diameter aspect ratio pore and machine learning-driven resistive pulse analyses. A high-spatiotemporal resolution of this electrical sensor enabled to observe galvanotactic response intrinsic to the microbes during their translocation. We demonstrated discrimination of the cellular motility via signal pattern classifications in a high-dimensional feature space. As the detection-to-decision can be completed within milliseconds, the present technique may be used for real-time screening of pathogenic bacteria for environmental and medical applications.

DOI： 10.1038/s41598-020-72508-3

researchmap

その他リンク： http://www.nature.com/articles/s41598-020-72508-3

掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Science and Business Media LLC  

<title>Abstract</title>Molecular dynamics (MD) simulation is used to analyze the mechanical properties of polymerized and nanoscale filled rubber. Unfortunately, the computation time for a simulation can require several months’ computing power, because the interactions of thousands of filler particles must be calculated. To alleviate this problem, we introduce a surrogate convolutional neural network model to achieve faster and more accurate predictions. The major difficulty when employing machine-learning-based surrogate models is the shortage of training data, contributing to the huge simulation costs. To derive a highly accurate surrogate model using only a small amount of training data, we increase the number of training instances by dividing the large-scale simulation results into 3D images of middle-scale filler morphologies and corresponding regional stresses. The images include fringe regions to reflect the influence of the filler constituents outside the core regions. The resultant surrogate model provides higher prediction accuracy than that trained only by images of the entire region. Afterwards, we extract the fillers that dominate the mechanical properties using the surrogate model and we confirm their validity using MD.

DOI： 10.1038/s41598-020-75038-0

researchmap

その他リンク： https://www.nature.com/articles/s41598-020-75038-0

掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Science and Business Media LLC  

DOI： 10.1038/s41598-020-68236-3

researchmap

その他リンク： http://www.nature.com/articles/s41598-020-68236-3

記述言語：英語   掲載種別：研究論文（学術雑誌）  

The variability of bioparticles remains a key barrier to realizing the competent potential of nanoscale detection into a digital diagnosis of an extraneous object that causes an infectious disease. Here, we report label-free virus identification based on machine-learning classification. Single virus particles were detected using nanopores, and resistive-pulse waveforms were analyzed multilaterally using artificial intelligence. In the discrimination, over 99% accuracy for five different virus species was demonstrated. This advance is accessed through the classification of virus-derived ionic current signal patterns reflecting their intrinsic physical properties in a high-dimensional feature space. Moreover, consideration of viral similarity based on the accuracies indicates the contributing factors in the recognitions. The present findings offer the prospect of a novel surveillance system applicable to detection of multiple viruses including new strains.

DOI： 10.1021/acssensors.0c01564

PubMed

researchmap

出版者・発行元：Research Square  

<title>Abstract</title>
High-throughput, high-accuracy detection of emerging viruses allows for pandemic prevention and control. Currently, reverse transcription-polymerase chain reaction (RT-PCR) is used to diagnose the presence of SARS-CoV-2. The principle of the test is to detect RNA in the virus using a pair of primers that specifically binds to the base sequence of the viral RNA. However, RT-PCR is a sophisticated technique requiring a time-consuming pretreatment procedure for extracting viral RNA from clinical specimens and to obtain<bold> </bold>high sensitivity. Here, we report a method for detecting novel coronaviruses with high sensitivity using artificial intelligent nanopores utilizing a simple procedure that does not require RNA extraction. Artificial intelligent nanopore platform consists of machine learning software on the servers, portable high-speed and high-precision current measuring instrument, and scalable, cost-effective semiconducting nanopore modules. Here we show that the artificial intelligent nanopores are successful in accurate identification of four types of coronaviruses, HCoV-229E, SARS-CoV, MERS-CoV, and SARS-CoV-2, which are usually extremely difficult to detect. The positive/negative diagnostics of the new coronavirus is achieved with a sensitivity of 95 % and specificity of 92 % with a 5-minute diagnosis. The platform enables high throughput diagnostics with low false negatives for the novel coronavirus.

DOI： 10.21203/rs.3.rs-97218/v1

researchmap

その他リンク： https://www.researchsquare.com/article/rs-97218/v1.html

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Royal Society of Chemistry  

The authors regret that there was an error in Fig. 2 in the original article. The scale for the flow rate in Fig. 2a was incorrect. The correct version of the figure is presented here. This does not affect the results or conclusions of the article. There were also some errors in the Supplementary Information. These have now been corrected in an updated version which is available online. (Figure presented) The Royal Society of Chemistry apologises for these errors and any consequent inconvenience to authors and readers.

DOI： 10.1039/d0cc90423f

Scopus

PubMed

researchmap

掲載種別：研究論文（学術雑誌）   出版者・発行元：American Chemical Society (ACS)  

DOI： 10.1021/acssensors.0c00845

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

PURPOSE: Glucose intolerance (GI), defined as either prediabetes or diabetes, promotes cardiovascular events in patients with myocardial infarction (MI). Using the pooled clinical data from patients with MI and GI in the completed ABC and PPAR trials, we aimed to identify their clinical risk factors for cardiovascular events. METHODS: Using the limitless-arity multiple testing procedure, an artificial intelligence (AI)-based data mining method, we analyzed 415,328 combinations of < 4 clinical parameters. RESULTS: We identified 242 combinations that predicted the occurrence of hospitalization for (1) percutaneous coronary intervention for stable angina, (2) non-fatal MI, (3) worsening of heart failure (HF), and (4) all causes, and we analyzed combinations in 1476 patients. Among these parameters, the use of proton pump inhibitors (PPIs) or plasma glucose levels > 200 mg/dl after 2 h of a 75 g oral glucose tolerance test were linked to the coronary events of (1, 2). Plasma BNP levels > 200 pg/dl were linked to coronary and cardiac events of (1, 2, 3). Diuretics use, advanced age, and lack of anti-dyslipidemia drugs were linked to cardiovascular events of (1, 3). All of these factors were linked to (4). Importantly, each finding was verified by independently drawn Kaplan-Meier curves, indicating that the determined factors accurately affected cardiovascular events. CONCLUSIONS: In most previous MI patients with GI, progression of GI, PPI use, or high plasma BNP levels were linked to the occurrence of coronary stenosis or recurrent MI. We emphasize that use of AI may comprehensively uncover the hidden risk factors for cardiovascular events.

DOI： 10.1007/s10557-020-06987-x

PubMed

researchmap

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr2009.html#abs-2009-12196

掲載種別：論文集(書籍)内論文  

© 2020, Springer Science+Business Media, LLC, part of Springer Nature. Super-resolution microscopy facilitates observation with an optical microscope at a higher spatial resolution than the diffraction limit of light; however, super-resolution observation with high biocompatibility remains challenging. Leading super-resolution techniques such as reversible saturable/switchable optical fluorescence transition (RESOLFT) and single-molecule localization microscopy (SMLM) need to illuminate a sample at an appreciably high power density of illumination, i.e., from 0.1 kW/cm2 to 1 GW/cm2. Unfortunately, that high power density gives rise to phototoxicity in live cells, and this may prevent widespread use of super-resolution imaging in the life sciences. In this study we show a technique of super-resolution imaging that can be performed at a very low power density of illumination, SPoD-OnSPAN (super-resolution polarization demodulation/on-state polarization angle narrowing). This achieves super-resolution observations at a power density as low as 1 W/cm2, and thereby high biocompatibility. The present technique is likely to be very useful for situations such as time-lapse super-resolution observations of live cells and tissues.

DOI： 10.1007/978-1-0716-0532-5_11

Scopus

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10618-019-00660-0

researchmap

掲載種別：研究論文（学術雑誌）  

© The Royal Society of Chemistry. A highly atom-economical enantioselective organocatalyzed Rauhut-Currier and [3+2] annulation sequence has been established by using a flow system. Suitable flow conditions were explored through reaction screening of multiple parameters using machine learning. Eventually, functionalized chiral spirooxindole analogues were obtained in high yield with good ee as a single diastereomer within one minute.

DOI： 10.1039/c9cc08526b

Scopus

PubMed

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：ACM  

DOI： 10.1145/3394486.3403062

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/kdd/kdd2020.html#TingXWZ20

掲載種別：研究論文（学術雑誌）  

Gas identification is one of the most important functions of a gas sensor system. To identify gas species from sensing signals without gas flow control such as pumps or mass flow controllers, it is necessary to extract decisive dynamic features from complex sensing signals due to uncontrolled airflow. For that purpose, various analysis methods using system identification techniques have been proposed, whereas a method that is not affected by a gas input pattern has been demanded to enhance the robustness of gas identification. Here we develop a novel gas identification protocol based on a transfer function ratio (TFR) that is intrinsically independent of a gas input pattern. By combining the protocol with MEMS-based sensors—Membrane-type Surface stress Sensors (MSS), we have realized gas identification with a free-hand measurement, in which one can simply hold a small sensor chip near samples. From sensing signals obtained through the free-hand measurement, we have developed highly accurate machine learning models that can identify odors of spices and herbs as well as solvent vapors. Since no bulky gas flow control units are required, this protocol will expand the applicability of gas sensors to portable electronics, leading to practical artificial olfaction.

DOI： 10.1038/s41598-019-46164-1

Scopus

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：NATURE PUBLISHING GROUP  

Gas identification is one of the most important functions of a gas sensor system. To identify gas species from sensing signals without gas flow control such as pumps or mass flow controllers, it is necessary to extract decisive dynamic features from complex sensing signals due to uncontrolled airflow. For that purpose, various analysis methods using system identification techniques have been proposed, whereas a method that is not affected by a gas input pattern has been demanded to enhance the robustness of gas identification. Here we develop a novel gas identification protocol based on a transfer function ratio (TFR) that is intrinsically independent of a gas input pattern. By combining the protocol with MEMS-based sensors-Membrane-type Surface stress Sensors (MSS), we have realized gas identification with a free-hand measurement, in which one can simply hold a small sensor chip near samples. From sensing signals obtained through the free-hand measurement, we have developed highly accurate machine learning models that can identify odors of spices and herbs as well as solvent vapors. Since no bulky gas flow control units are required, this protocol will expand the applicability of gas sensors to portable electronics, leading to practical artificial olfaction.

DOI： 10.1038/s41598-019-46164-1

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1021/acsomega.9b00946

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1021/acs.jpcc.9b03908

researchmap

掲載種別：研究論文（学術雑誌）  

© 2019 The Royal Society of Chemistry. We propose an artificial intelligence-based chemical-sensing system integrating a porous gate field-effect transistor (PGFET) array modified by gas chromatography stationary phase materials and machine-learning techniques. The chemically sensitive PGFET array generates cross-reactive signals for computational analysis and shows potential for applications to compact intelligent sensing devices, including mobile electronic noses.

DOI： 10.1039/c8me00097b

Scopus

researchmap

記述言語：日本語   出版者・発行元：公益社団法人 応用物理学会  

DOI： 10.11470/jsapmeeting.2019.1.0_2513

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1039/C8NR09042D

Web of Science

researchmap

掲載種別：研究論文（国際会議プロシーディングス）  

© 2019 CBMS-0001. As the WHO delivers a strong warning about drug-resistant bacteria, their types are increasing along with improper use of antibiotic drugs by medical professions, due to the fact that conventional methods cannot identify drug-resistant bacteria in a short time, and show a guideline for proper use of antibiotic drugs rapidly. Herein, we demonstrated a rational methodology for drug-resistant bacteria identification via machine learning analysis on fine current changes from bacteria inner ion leakages, which given by highly applied electric fields in microchannels We believe that our methodology opens up a new way for proper use of antibiotic drugs against drug-resistant bacteria.

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：PMLR  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/acml/2019

掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/ISOEN.2019.8823511

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/isoen/isoen2019.html#ShunoriYWHNNOOW19

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1902.html#abs-1902-03402

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1039/C8LC01371C

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1039/C8NR08632J

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10994-018-5737-x

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.7717/peerj-cs.169

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/peerj-cs/peerj-cs5.html#BlobaumJWSS19

掲載種別：研究論文（国際会議プロシーディングス）  

© 2019 CBMS-0001. In this work, novel system of multiple viral discrimination via resistive pulse sensing is proposed. Utilizing artificial intelligence (AI) to classify ionic current signals measured by nanopore, >70% accuracy was demonstrated among 5 species of respiratory tract infectious viruses at a single particle level. This technique can be used as a useful point-of-care testing tool for diagnosis of infectious diseases.

Scopus

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1021/ACSSENSORS.9B00113

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1907.html#abs-1907-01104

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Immunosensing is a bioanalytical technique capable of selective detections of pathogens by utilizing highly specific and strong intermolecular interactions between recognition probes and antigens. Here, we exploited the molecular mechanism in artificial nanopores for selective single-virus identifications. We designed hemagglutinin antibody mimicking oligopeptides with a weak affinity to influenza A virus. By functionalizing the pore wall surface with the synthetic peptides, we rendered specificity to virion-nanopore interactions. The ligand binding thereof was found to perturb translocation dynamics of specific viruses in the nanochannel, which facilitated digital typing of influenza by the resistive pulse bluntness. As amino acid sequence degrees of freedom can potentially offer variety of recognition ability to the molecular probes, this peptide nanopore approach can be used as a versatile immunosensor with single-particle sensitivity that promises wide applications in bioanalysis including bacterial and viral screening to infectious disease diagnosis.

DOI： 10.1021/jacs.8b10854

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Rapid diagnosis of flu before symptom onsets can revolutionize our health through diminishing a risk for serious complication as well as preventing infectious disease outbreak. Sensor sensitivity and selectivity are key to accomplish this goal as the number of virus is quite small at the early stage of infection. Here we report on label-free electrical diagnostics of influenza based on nanopore analytics that distinguishes individual virions by their distinct physical features. We accomplish selective resistive-pulse sensing of single flu virus having negative surface charges in a physiological media by exploiting electroosmotic flow to filter contaminants at the Si3N4 pore orifice. We demonstrate identifications of allotypes with 68% accuracy at the single-virus level via pattern classifications of the ionic current signatures. We also show that this discriminability becomes >95% under a binomial distribution theorem by ensembling the pulse data of >20 virions. This simple mechanism is versatile for point-of-care tests of a wide range of flu types.

DOI： 10.1038/s41598-018-34665-4

PubMed

researchmap

記述言語：日本語   出版者・発行元：公益社団法人 応用物理学会  

DOI： 10.11470/jsapmeeting.2018.2.0_2648

researchmap

掲載種別：研究論文（学術雑誌）  

Nanomechanical sensors - emerging chemical sensors which detect changes in mechanical properties caused by gas sorption - have been attracting much attention owing to their high sensitivity and versatility. In the data analysis of sensing signals, empirically extracted signal features have been commonly employed to identify the gas species. Such an empiric approach cannot be optimized further without a solid guideline, resulting in a limited identification performance. Therefore, a new analytical protocol based on intrinsic physical properties of a target gas and a receptor material has been highly demanded. In this study, we have developed a parameter estimation protocol based on a theoretical model for a cantilever-type nanomechanical sensor coated with a viscoelastic material. This protocol provides a practical estimation method for intrinsic parameters, which can be used for gas identification. As a demonstration of gas identification based on intrinsic parameters, we focused on the time constant for gas diffusion τs, which reflects the physicochemical interaction between gas species and a receptor material. Based on τs estimated from different receptor materials, we succeeded in the identification of solvent vapors. This parameter estimation protocol not only enables the gas identification based on the intrinsic property of target gases, but also provides a scientific guideline for the selection and optimization of receptor materials for nanomechanical sensors.

DOI： 10.1063/1.5036686

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：AMER INST PHYSICS  

Nanomechanical sensors-emerging chemical sensors which detect changes in mechanical properties caused by gas sorption-have been attracting much attention owing to their high sensitivity and versatility. In the data analysis of sensing signals, empirically extracted signal features have been commonly employed to identify the gas species. Such an empiric approach cannot be optimized further without a solid guideline, resulting in a limited identification performance. Therefore, a new analytical protocol based on intrinsic physical properties of a target gas and a receptor material has been highly demanded. In this study, we have developed a parameter estimation protocol based on a theoretical model for a cantilever-type nanomechanical sensor coated with a viscoelastic material. This protocol provides a practical estimation method for intrinsic parameters, which can be used for gas identification. As a demonstration of gas identification based on intrinsic parameters, we focused on the time constant for gas diffusion tau(s), which reflects the physicochemical interaction between gas species and a receptor material. Based on tau(s) estimated from different receptor materials, we succeeded in the identification of solvent vapors. This parameter estimation protocol not only enables the gas identification based on the intrinsic property of target gases, but also provides a scientific guideline for the selection and optimization of receptor materials for nanomechanical sensors. (C) 2018 Author(s).

DOI： 10.1063/1.5036686

Web of Science

researchmap

DOI： 10.1093/jmicro/dfy004

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Bioinspired pore sensing for selective detection of flagellated bacteria was investigated. The Au micropore wall surface was modified with a synthetic peptide designed from toll-like receptor 5 (TLR5) to mimic the pathogen-recognition capability. We found that intermolecular interactions between the TLR5-derived recognition peptides and flagella induce ligand-specific perturbations in the translocation dynamics of Escherichia coli, which facilitated the discrimination between the wild-type and flagellin-deletion mutant (ΔfliC) by the resistive pulse patterns thereby demonstrating the sensing of bacteria at a single-cell level. These results provide a novel concept of utilizing weak intermolecular interactions as a recognition probes for single-cell microbial identification.

DOI： 10.1021/acs.analchem.7b04950

Web of Science

Scopus

PubMed

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1021/ACSSENSORS.8B01214

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1802.html#abs-1802-06698

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1812.html#abs-1812-03395

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SIAM  

DOI： 10.1137/1.9781611975321.64

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/sdm/sdm2018.html#KisamoriWKF18

掲載種別：研究論文（学術雑誌）  

DOI： 10.1021/ACSAMI.8B11819

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10994-017-5693-x

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：PMLR  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/aistats/2018

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDM.2018.00015

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2018.html#TingAW18

掲載種別：研究論文（国際会議プロシーディングス）  

We report on a use of low thickness-to-diameter aspect-ratio pores and machine learning for discriminating single-particles. In stark contrast to the conventional resistive pulse analysis that looks only on the height of the ionic current spike signals obtained with long fluidic channels to deduce the particle size, the present technique goes multimodal with state-of-the-art machine learning algorithms to extract and compare multiple features of each electrical signatures in a low-aspect-ratio pore sensor whereby offering a way to identify micro- and nano-objects by their multiplex physical properties.

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Conventional concepts of resistive pulse analysis is to discriminate particles in liquid by the difference in their size through comparing the amount of ionic current blockage. In sharp contrast, we herein report a proof-of-concept demonstration of the shape sensing capability of solid-state pore sensors by leveraging the synergy between nanopore technology and machine learning. We found ionic current spikes of similar patterns for two bacteria reflecting the closely resembled morphology and size in an ultra-low thickness-to-diameter aspect-ratio pore. We examined the feasibility of a machine learning strategy to pattern-analyse the sub-nanoampere corrugations in each ionic current waveform and identify characteristic electrical signatures signifying nanoscopic differences in the microbial shape, thereby demonstrating discrimination of single-bacterial cells with accuracy up to 90%. This data-analytics-driven microporescopy capability opens new applications of resistive pulse analyses for screening viruses and bacteria by their unique morphologies at a single-particle level.

DOI： 10.1038/s41598-017-17443-6

PubMed

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10115-017-1046-0

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10994-016-5586-4

researchmap

掲載種別：研究論文（国際会議プロシーディングス）  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/esann/2017

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

DOI： 10.1109/DSAA.2017.28

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dsaa/dsaa2017.html#WashioIY17

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：AMER PHYSICAL SOC  

The accurate detection of small deviations in given density matrice is important for quantum information processing, which is a difficult task because of the intrinsic fluctuation in density matrices reconstructed using a limited number of experiments. We previously proposed a method for decoherence error detection using a machine-learning technique [S. Hara, T. Ono, R. Okamoto, T. Washio, and S. Takeuchi, Phys. Rev. A 89, 022104 (2014)]. However, the previous method is not valid when the errors are just changes in phase. Here, we propose a method that is valid for arbitrary errors in density matrices. The performance of the proposed method is verified using both numerical simulation data and real experimental data.

DOI： 10.1103/PhysRevA.94.042341

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：AMER CHEMICAL SOC  

Resistive pulse sensing with nanopores having a low thickness to -diameter aspect-ratio structure is expected to enable high-spatial-resolution analysis of nanoscale objects in a liquid. Here we investigated the sensing capability of low-aspect-ratio pore sensors by monitoring the ionic current blockades during translocation of polymeric nanobeads. We detected numerous small current spikes due to partial occlusion of the pore orifice by particles diffusing therein reflecting the expansive electrical sensing zone of the low-aspect-ratio pores. We also found wide variations in the ion current line-shapes in the particle capture stage suggesting random incident angle of the particles drawn into the pore. In sharp contrast, the ionic profiles were highly reproducible in the post-translocation regime by virtue of the spatial confinement in the pore that effectively constricts the stochastic capture dynamics into a well-defined ballistic motion. These results, together with multiphysics simulations, indicate that the resistive pulse height is highly dependent on the nanoscopic single-particle trajectories involved in ultrathin pore sensors. The present finding indicates the importance of regulating the translocation pathways of analytes in low-aspect-ratio pores for improving the discriminability toward single-bioparticle tomography in liquid.

DOI： 10.1021/acsnano.5b05906

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1145/2710025

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-319-31753-3

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-319-31750-2

researchmap

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1610.html#BlobaumWS16

記述言語：英語   出版者・発行元：一般社団法人 日本毒性学会  

Structured regularization is a mathematical technique which incorporates prior structural knowledge among variables into regression analysis to make a sparse estimation reflecting relationships among them. Abundance of structural information in biology, such as pathways formed by genes, transcripts, and proteins, especially suits well its application. Previously, we reported on the first application of latent group Lasso, a group-based regularization method, in toxicogenomics, with genes regulated by the same transcription factor treated as a group. We revealed that it achieved good predictive performances comparable to Lasso and allowed us to discuss mechanisms behind liver weight gain in rats based on selected groups. Latent group Lasso, however, does not lead to a sparse estimation, due to large group sizes in our analytical setting. In this study, we applied graph-based regularization methods, generalized fused Lasso and graph Lasso, for the same data, with regulatory networks formed by transcription factors and their downstream genes as a graph. These methods are expected to make a sparser estimation since they select variables based on edges. Comparisons showed that graph Lasso generated an accurate, biologically relevant and sparse model that could not have been possible with latent group Lasso and generalized fused Lasso.

DOI： 10.2131/fts.3.39

researchmap

その他リンク： http://search.jamas.or.jp/link/ui/2017038113

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s41060-016-0035-9

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer Verlag  

Having knowledge about the real underlying causal structure of a data generation process has various implications for different machine learning problems. We address the idea of causal and anticausal learning with respect to a comparison of discriminative and generative models. In particular, we conjecture the hypothesis that generative models perform better in anticausal problems than in causal problems. We empirical evaluate our hypothesis with different real-world data sets.

DOI： 10.1007/978-3-319-28379-1_15

Scopus

researchmap

記述言語：英語   出版者・発行元：一般社団法人 日本毒性学会  

Regression analysis such as linear regression and logistic regression has often been employed to construct toxicogenomic predictive models, which forecast toxicological effects of chemical compounds in human or animals based on gene expression data. While in general these techniques can generate an accurate and sparse model when a regularization term is added to a loss function, they ignore structural relationships behind genes which form vast regulatory networks and interact with each other. Recently, several reports proposed structured sparsity-inducing norms to incorporate prior structural information and make a model reflecting relationships between variables. In this study, assuming that genes regulated by the same transcription factor should be selected together, we applied the latent group Lasso technique on toxicogenomic data with transcription factor networks as prior knowledge. We compared generated classifiers for liver weight gain in rats between the latent group Lasso and Lasso. The latent group Lasso was comparable or superior to the Lasso in terms of predictive performances (balanced accuracy: 74% vs. 72%, sensitivity: 62% vs. 62%, specificity: 86% vs. 83%). Besides, groups selected by the latent group Lasso suggested involvement of Wnt/β-catenin signaling pathway. Such mechanism-related analysis could not have been possible with the Lasso and is one of the advantages of the latent group Lasso.

DOI： 10.2131/fts.2.161

researchmap

その他リンク： http://search.jamas.or.jp/link/ui/2016381691

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10994-015-5524-x

researchmap

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Institute of Electronics Information Communication Engineers  

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Elsevier Inc.  

While the recent advent of new technologies in biology such as DNA microarray and next-generation sequencer has given researchers a large volume of data representing genome-wide biological responses, it is not necessarily easy to derive knowledge that is accurate and understandable at the same time. In this study, we applied the Classification Based on Association (CBA) algorithm, one of the class association rule mining techniques, to the TG-GATEs database, where both toxicogenomic and toxicological data of more than 150 compounds in rat and human are stored. We compared the generated classifiers between CBA and linear discriminant analysis (LDA) and showed that CBA is superior to LDA in terms of both predictive performances (accuracy: 83% for CBA vs. 75% for LDA, sensitivity: 82% for CBA vs. 72% for LDA, specificity: 85% for CBA vs. 75% for LDA) and interpretability.

DOI： 10.1016/j.toxrep.2014.10.014

Scopus

researchmap

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

DOI： 10.11517/jsaifpai.94.0_09

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I025861071

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1103/PhysRevA.89.022104

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1162/NECO_a_00533

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/neco/neco26.html#TashiroSHW14

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDM.2014.33

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2014.html#AryalTHW14

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-319-06605-9_42

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd2014-2.html#AryalTWW14

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1401.html#InazumiWSSYK14

掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.patcog.2014.01.013

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：PERGAMON-ELSEVIER SCIENCE LTD  

Properties of data are frequently seen to vary depending on the sampled situations, which usually change along a time evolution or owing to environmental effects. One way to analyze such data is to find invariances, or representative features kept constant over changes. The aim of this paper is to identify one such feature, namely interactions or dependencies among variables that are common across multiple datasets collected under different conditions. To that end, we propose a common substructure learning (CSSL) framework based on a graphical Gaussian model. We further present a simple learning algorithm based on the Dual Augmented Lagrangian and the Alternating Direction Method of Multipliers. We confirm the performance of CSSL over other existing techniques in finding unchanging dependency structures in multiple datasets through numerical simulations on synthetic data and through a real world application to anomaly detection in automobile sensors. (c) 2012 Elsevier Ltd. All rights reserved.

DOI： 10.1016/j.neunet.2012.11.004

Web of Science

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/nn/nn38.html#HaraW13

DOI： 10.1527/tjsai.28.13

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDMW.2013.17

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdmw2013.html#DemeshkoWK13

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10115-013-0612-3

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.neunet.2013.05.007

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/nn/nn46.html#SogawaUKW13

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：ACM  

DOI： 10.1145/2487575.2488211

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/kdd/kdd2013.html#KengneFTIRWS13

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

Recently, there is a growing need for statistical learning of causal structures in data with many variables. A structural equation model called Linear Non-Gaussian Acyclic Model (LiNGAM) has been extensively studied to uniquely estimate causal structures in data. The key assumptions are that external influences are independent and follow non-Gaussian distributions. However, LiNGAM does not capture temporal structural changes in observed data. In this paper, we consider learning causal structures in longitudinal data that collects samples over a period of time. In previous studies of LiNGAM, there was no model specialized to handle longitudinal data with multiple samples. Therefore, we propose a new model called longitudinal LiNGAM and a new estimation method using the information on temporal structural changes and non-Gaussianity of data. The new approach requires less assumptions than previous methods.

DOI： 10.1109/MLSP.2013.6661912

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：American Nuclear Society  

Scopus

researchmap

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

近年，膨大な観測データに潜む因果構造を推定するための統計的手法が必要とされている．本研究では経時データと呼ばれる，時間の推移とともに観測されたデータを用いた因果構造の推定法を提案する．経時データの性質とデータの非ガウス性を利用することで、従来よりも緩い仮定の下で因果構造推定が可能である．更に推定された因果構造に対して時間の経過とともに有意な変化をしている部分を統計的に検出する方法も提案する．

DOI： 10.11517/pjsai.jsai2013.0_3d14

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I025960543

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

近年様々な分野で大量の観測データが蓄積されており，因果分析法に対するニーズは高まっている。最近の研究により，データの非ガウス性を利用することで変数間の因果的順序を同定できる場合があることがわかっている。本研究では未観測交絡変数がある場合にその値を離散化してベイズ推定を行い，二変数間の因果順序を同定する手法を提案する。これにより，潜在交絡変数が存在しても頑健な推定することが可能となる。

DOI： 10.11517/pjsai.jsai2013.0_3d15

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I025962897

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：PERGAMON-ELSEVIER SCIENCE LTD  

Non-stationary effects are ubiquitous in real world data. In many settings, the observed signals are a mixture of underlying stationary and non-stationary sources that cannot be measured directly. For example, in EEG analysis, electrodes on the scalp record the activity from several sources located inside the brain, which one could only measure invasively. Discerning stationary and non-stationary contributions is an important step towards uncovering the mechanisms of the data generating system. To that end, in Stationary Subspace Analysis (SSA), the observed signal is modeled as a linear superposition of stationary and non-stationary sources, where the aim is to separate the two groups in the mixture. In this paper, we propose the first SSA algorithm that has a closed form solution. The novel method, Analytic SSA (ASSA), is more than 100 times faster than the state-of-the-art, numerically stable, and guaranteed to be optimal when the covariance between stationary and non-stationary sources is time-constant. In numerical simulations on wide range of settings, we show that our method yields superior results, even for signals with time-varying group-wise covariance. In an application to geophysical data analysis, ASSA extracts meaningful components that shed new light on the Pi 2 pulsations of the geomagnetic field. (c) 2012 Elsevier Ltd. All rights reserved.

DOI： 10.1016/j.neunet.2012.04.001

Web of Science

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/nn/nn33.html#HaraKWBTY12

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：The Institute of Electronics, Information and Communication Engineers  

The mining of frequent subgraphs from labeled graph data has been studied extensively. Furthermore, much attention has recently been paid to frequent pattern mining from graph sequences. A method, called GTRACE, has been proposed to mine frequent patterns from graph sequences under the assumption that changes in graphs are gradual. Although GTRACE mines the frequent patterns efficiently, it still needs substantial computation time to mine the patterns from graph sequences containing large graphs and long sequences. In this paper, we propose a new version of GTRACE that permits efficient mining of frequent patterns based on the principle of a reverse search. The underlying concept of the reverse search is a general scheme for designing efficient algorithms for hard enumeration problems. Our performance study shows that the proposed method is efficient and scalable for mining both long and large graph sequence patterns and is several orders of magnitude faster than the original GTRACE.

DOI： 10.1587/transinf.E95.D.1947

CiNii Books

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：The Institute of Electronics, Information and Communication Engineers  

The mining of a complete set of frequent subgraphs from labeled graph data has been studied extensively. Furthermore, much attention has recently been paid to frequent pattern mining from graph sequences (dynamic graphs or evolving graphs). In this paper, we define a novel subgraph subsequence class called an "induced subgraph subsequence" to enable the efficient mining of a complete set of frequent patterns from graph sequences containing large graphs and long sequences. We also propose an efficient method for mining frequent patterns, called "FRISSs (<u>F</u>requent <u>R</u>elevant, and <u>I</u>nduced <u>S</u>ubgraph <u>S</u>ubsequences)", from graph sequences. The fundamental performance of the method is evaluated using artificial datasets, and its practicality is confirmed through experiments using a real-world dataset.

DOI： 10.1587/transinf.E95.D.1590

CiNii Books

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1002/sam.11136

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-642-33269-2_62

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/icann/2012-1

掲載種別：研究論文（国際会議プロシーディングス）  

researchmap

その他リンク： http://papers.nips.cc/paper/4815-weighted-likelihood-policy-search-with-model-selection

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SPRINGER-VERLAG BERLIN  

Sparse Inverse Covariance Selection (SICS) is a popular tool identifying an intrinsic relationship between continuous random variables. In this paper, we treat the extension of SICS to the grouped feature model in which the state-of-the-art SICS algorithm is no longer applicable. Such an extended model is essential when we aim to find a group-wise relationships between sets of variables, e. g. unknown interactions between groups of genes. We tackle the problem with a technique called Dual Augmented Lagrangian (DAL) that provides an efficient method for grouped sparse problems. We further improve the DAL framework by combining the Alternating Direction Method of Multipliers (ADMM), which dramatically simplifies the entire procedure of DAL and reduce the computational cost. We also provide empirical comparisons of the proposed DAL-ADMM algorithm against existing methods.

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

We propose a method to extract row/column-wise heterogeneous elements between two precision matrices for an anomaly localization. We formulate the task as a convex optimization problem using a regularizaion term that penalizes row/column-wise differences between two matrices. The fundamental difficulties of the problem are that the proposed regularization term (1) is a sum of group-wise regularizations with overlapping supports between the groups, (2) penalizes matrices in a symmetric manner. Our proposed algorithm with an alternating direction method of multipliers can deal with these two difficulties efficiently resulting in a very simple formulation with each updating step computed analytically. We also show the validity of the proposed method through an anomaly localization simulation using a real world data.

DOI： 10.1109/ICDMW.2012.10

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1587/transinf.E95.D.2599

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/ieicet/ieicet95d.html#KuwajimaWL12

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-642-32695-0_14

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pricai/pricai2012.html#InokuchiYWMAIK12

掲載種別：研究論文（学術雑誌）   出版者・発行元：AUAI Press  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1202.html#abs-1202-3736

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

We have been considering a problem of finding significant connection strengths of variables in a linear non-Gaussian causal model called LiNGAM. In our previous work, bootstrap confidence intervals of connection strengths were simultaneously computed in order to test their statistical significance. However, the distribution of estimated elements in an adjacency matrix obtained by the bootstrap method was not close enough to the real distribution even though the number of bootstrap replications was increased. Moreover, such a naive approach raised the multiple comparison problem which many directed edges were likely to be falsely found significant. In this study, we propose a new approach used to correct the distribution obtained by the bootstrap method. We also apply a representative technique of multiple comparison, the Bonferroni correction, then evaluate its performance. The result of this study shows that the new distribution is more stable and also even closer to the real distribution. Besides, the number of falsely found significant edges is less than the previous approach.

DOI： 10.1109/ICDMW.2012.134

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SPRINGER-VERLAG BERLIN  

The performance of active learning (AL) is crucially influenced by the existence of outliers in input samples. In this paper, we propose a robust pool-based AL measure based on the density power divergence. It is known that the density power divergence can be accurately estimated even under the existence of outliers within data. We further derive an AL scheme based on an asymptotic statistical analysis on the M-estimator. The performance of the proposed framework is investigated empirically using artificial and real-world data.

DOI： 10.1007/978-3-642-34487-9_72

Web of Science

researchmap

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

近年様々な分野で大量の観測データが蓄積されており，データ生成過程の推定手法に対するニーズは高まっている。最近の研究により，データの非ガウス性を利用することでデータ生成過程を同定できる場合があることがわかっている。本研究ではデータ生成過程の推定と同時にモデルがデータに適合しているかを確認する手法を提案する。これにより，モデル適合部分についてのみデータ生成過程を推定することが可能となる。

DOI： 10.11517/pjsai.jsai2012.0_4b1r26

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I025970995

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：PERGAMON-ELSEVIER SCIENCE LTD  

Many statistical methods have been proposed to estimate causal models in classical situations with fewer variables than observations. However, modern datasets including gene expression data increase the needs of high-dimensional causal modeling in challenging situations with orders of magnitude more variables than observations. In this paper, we propose a method to find exogenous variables in a linear non-Gaussian causal model, which requires much smaller sample sizes than conventional methods and works even under orders of magnitude more variables than observations. Exogenous variables work as triggers that activate causal chains in the model, and their identification leads to more efficient experimental designs and better understanding of the causal mechanism. We present experiments with artificial data and real-world gene expression data to evaluate the method. (C) 2011 Elsevier Ltd. All rights reserved.

DOI： 10.1016/j.neunet.2011.05.017

Web of Science

PubMed

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/nn/nn24.html#SogawaSSHWI11

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

定常な多変数時系列データにARMAモデルを適用し、過去変数からの影響を解析する手法はよく知られている。近年、ARMAモデルと構造方程式を組み合わせることで同時変数間の影響も説明可能としたARMA-LiNGAMモデルが提案された。本論文ではデータの非ガウス性を利用したARMA-LiNGAMモデルの推定手法を説明し、従来の推定手法の問題点を指摘したうえでその問題点を解決する推定手法を提案する。

DOI： 10.11517/pjsai.jsai2011.0_2e35

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I025900145

掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.neucom.2011.02.008

researchmap

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/jmlr/jmlr12.html#ShimizuISHKWHB11

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDM.2011.47

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2011.html#TingWWL11

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1108.html#abs-1108-4217

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SPRINGER-VERLAG BERLIN  

Learning underlying mechanisms of data generation is of great interest in the scientific and engineering fields amongst others. Finding dependency structures among variables in the data is one possible approach for the purpose, and is an important task in data mining. In this paper, we focus on learning dependency substructures shared by multiple datasets. In many scenarios, the nature of data varies due to a change in the surrounding conditions or non-stationary mechanisms over the multiple datasets. However, we can also assume that the change occurs only partially and some relations between variables remain unchanged. Moreover, we can expect that such commonness over the multiple datasets is closely related to the invariance of the underlying mechanism. For example, errors in engineering systems are usually caused by faults in the sub-systems with the other parts remaining healthy. In such situations, though anomalies are observed in sensor values, the underlying invariance of the healthy sub-systems is still captured by some steady dependency structures before and after the onset of the error. We propose a structure learning algorithm to find such invariances in the case of Graphical Gaussian Models (GGM). The proposed method is based on a block coordinate descent optimization, where subproblems can be solved efficiently by existing algorithms for Lasso and the continuous quadratic knapsack problem. We confirm the validity of our approach through numerical simulations and also in applications with real world datasets extracted from the analysis of city-cycle fuel consumption and anomaly detection in car sensors.

DOI： 10.1007/978-3-642-23783-6_1

Web of Science

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pkdd/pkdd2011-2.html#HaraW11

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/ICDMW.2011.30

Scopus

researchmap

出版者・発行元：AUAI Press  

researchmap

その他リンク： http://dblp.uni-trier.de/db/conf/uai/uai2011.html#conf/uai/InazumiWSSYK11

掲載種別：研究論文（学術雑誌）  

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

Many multivariate Gaussianity-based techniques for identifying causal networks of observed variables have been proposed. These methods have several problems such that they cannot uniquely identify the causal networks without any prior knowledge. To alleviate this problem, a non-Gaussianity-based identification method LiNGAM was proposed. Though the LiNGAM potentially identifies a unique causal network without using any prior knowledge, it needs to properly examine independence assumptions of the causal network and search the correct causal network by using finite observed data points only. On another front, a kernel based independence measure that evaluates the independence more strictly was recently proposed. In addition, some advanced generic search algorithms including beam search have been extensively studied in the past. In this paper, we propose some variants of the LiNGAM method which introduce the kernel based method and the beam search enabling more accurate causal network identification. Furthermore, we experimentally characterize the LiNGAM and its variants in terms of accuracy and robustness of their identification.

Web of Science

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10115-010-0346-4

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1504/IJKESDP.2010.037492

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1504/IJAIP.2010.030538

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1587/transinf.E93.D.2792

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/ieicet/ieicet93d.html#InokuchiW10

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/corr/corr1006.html#abs-1006-5041

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SIAM  

DOI： 10.1137/1.9781611972801.41

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/sdm/sdm2010.html#InokuchiW10

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-642-13672-6_18

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd2010-2.html#InokuchiW10

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-642-16184-1_15

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis2010.html#VinhIW10

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SPRINGER-VERLAG BERLIN  

We discuss causal structure learning based on linear structural equation models. Conventional learning methods most often assume Gaussianity and create many indistinguishable models. Therefore, in many cases it is difficult to obtain much information on the structure. Recently, a non-Gaussian learning method called LiNGAM has been proposed to identify the model structure without using prior knowledge on the structure. However, more efficient learning can be achieved if some prior knowledge on a part of the structure is available. In this paper, we propose to use prior knowledge to improve the performance of a state-of-art non-Gaussian method. Experiments on artificial data show that the accuracy and computational time are significantly improved even if the amount of prior knowledge is not so large.

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SPRINGER-VERLAG BERLIN  

Understanding non-stationary effects is one of the key challenges in data analysis. However, in many settings the observation is a mixture of stationary and non-stationary sources. The aim of Stationary Subspace Analysis (SSA) is to factorize multivariate data into its stationary and non-stationary components. In this paper, we propose a novel SSA algorithm (ASSA) that extracts stationary sources from multiple time series blocks. It has a globally optimal solution under certain assumptions that can be obtained by solving a generalized eigenvalue problem. Apart from the numerical advantages, we also show that compared to the existing method, fewer blocks are required in ASSA to guarantee the identifiability of the solution. We demonstrate the validity of our approach in simulations and in an application to domain adaptation.

DOI： 10.1007/978-3-642-17537-4_52

Web of Science

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/iconip/iconip2010-1.html#HaraKWB10

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

The objective of this paper is to introduce further development of a data mining tool, MUSASHI (Mining Utilities and System Architecture for Scalable processing of HIstorical data), for service computing. Recent advances in information systems have allowed us to gather enormous amounts of data on marketing. However, these gathered data have been individually stored at each company, and have never been integrated because of a lack of techniques to analyze the data in an integrated way and to handle the large amount of data efficiently. To address this issue, we are currently investigating a way to provide a data mining platform as a service so that users can apply various data mining techniques to their marketing data with ease and at a low cost. For this purpose, we have developed an ASP platform leveraging distributed computing technology rep- resented by Cloud computing. This paper describes the ASP platform for data mining services and introduces an empirical application of data mining using our platform. © 2010 IEEE.

DOI： 10.1109/GrC.2010.168

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/grc/grc2010.html#IchikawaYW10

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

Many statistical methods have been proposed to estimate causal models in
classical situations with fewer variables than observations (p<n, p: the number
of variables and n: the number of observations). However, modern datasets
including gene expression data need high-dimensional causal modeling in
challenging situations with orders of magnitude more variables than
observations (p>>n). In this paper, we propose a method to find exogenous
variables in a linear non-Gaussian causal model, which requires much smaller
sample sizes than conventional methods and works even when p>>n. The key idea
is to identify which variables are exogenous based on non-Gaussianity instead
of estimating the entire structure of the model. Exogenous variables work as
triggers that activate a causal chain in the model, and their identification
leads to more efficient experimental designs and better understanding of the
causal mechanism. We present experiments with artificial data and real-world
gene expression data to evaluate the method.

DOI： 10.1007/978-3-642-15819-3_10

Scopus

arXiv

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/icann/2010-1

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

構造方程式モデルと呼ばれる統計モデルを用いた因果構造探索を議論する。従来法はデータ共分散行列の情報しか用いないことが多いため、モデルを一意に同定できない問題があった。最近、データの非正規性を活用することで、この問題を解消できる場合があることがわかってきた。だが、現在の非正規性を使う方法は、データ情報のみを用い、事前情報があっても利用しない。本発表では、事前情報も利用することで探索性能の向上を図る。

DOI： 10.11517/pjsai.jsai2010.0_1a53

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I025892825

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SPRINGER-VERLAG BERLIN  

This research aims to present an analysis to optimally allocate advertising budgets based on single source data on consumers' views of TV advertising. A model of consumer behavior and an optimality criterion for the advertising budget allocation are proposed together with a CA based optimization algorithm. Through the analysis, we discovered some knowledge to improve the effectiveness of advertising for several products.

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

Given a set of graphs with class labels, discriminative subgraphs appearing therein are useful to construct a classification model. A graph mining technique called Chunkingless Graph-Based Induction (Cl-GBI) can find such discriminative subgraphs from graph structured data. But, it sometimes happens that Cl-GBI cannot extract subgraphs that are good enough to characterize the given data due to its time and space complexities. Thus, to improve its efficiency, we propose pruning methods based on the upper-bound of information gain that is used as a criterion for discriminability of subgraphs in Cl-GBI. The upper-bound of information gain of a subgraph is the maximal one that its super graph can achieve. By comparing the upper-bound of each subgraph with the best information gain at the moment, Cl-GBI can exclude unfruitful subgraphs from its search space. Furthermore, we experimentally evaluate the effectiveness of the pruning methods on a real world and artificial datasets. © Springer-Verlag Berlin Heidelberg 2009.

DOI： 10.1007/978-3-642-01715-5_5

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）  

Structural equation models and Bayesian networks have been widely used to analyze causal relations between continuous variables. In such frameworks, linear acyclic models are typically used to model the datagenerating process of variables. Recently, it was shown that use of non-Gaussianity identifies a causal ordering of variables in a linear acyclic model without using any prior knowledge on the network structure, which is not the case with conventional methods. However, existing estimation methods are based on iterative search algorithms and may not converge to a correct solution in a finite number of steps. In this paper, we propose a new direct method to estimate a causal ordering based on non-Gaussianity. In contrast to the previous methods, our algorithm requires no algorithmic parameters and is guaranteed to converge to the right solution within a small fixed number of steps if the data strictly follows the model.

Scopus

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s00354-009-0065-0

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s12626-008-0005-3

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-642-05224-8

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-642-00399-8

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

This research aims to present an analysis to optimally allocate advertising budgets based on single source data on consumers' views of TV advertising. A model of consumer behavior and an optimality criterion for the advertising budget allocation are proposed together with a GA based optimization algorithm. Through the analysis, we discovered some knowledge to improve the effectiveness of advertising for several products. © 2009 Springer Berlin Heidelberg.

DOI： 10.1007/978-3-642-04592-9_34

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/kes/kes2009-2.html#IchikawaYNW09

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-540-68125-0

researchmap

掲載種別：研究論文（学術雑誌）  

DOI： 10.1109/TKDE.2007.190695

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDM.2008.106

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2008.html#InokuchiW08

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-540-85565-1_87

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/kes/kes2008-2.html#YadaWUN08

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/ida/ida12.html#NguyenW08

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SIAM  

DOI： 10.1137/1.9781611972788.35

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/sdm/sdm2008.html#KidoKW08

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s00354-007-0015-7

researchmap

出版者・発行元：Springer  

DOI： 10.1007/978-3-540-73920-3_5

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/book2007.html#WashioM07

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：SPRINGER-VERLAG BERLIN  

The knowledge on the relation between a financial state of an enterprise and its future profit will efficiently and securely reduce the negative risk and increase the positive risk on the decision making needed in the management of the enterprise and the investment in stock markets. Generally speaking, the relation is considered to have a highly complicated structure containing the influences from various financial factors characterizing the enterprise. Recent development of data mining techniques has significantly extended the power to model such a complicated relation in accurate and tractable manners. In this study, we assessed the feasibility to model the relation in the framework of data mining, and analyzed the characteristics of the model.

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer Verlag  

A machine learning technique called Graph-Based Induction (GBI) efficiently extracts typical patterns from graph-structured data by stepwise pair expansion (pairwise chunking). It is very efficient because of its greedy search. We have expanded GBI to construct a decision tree that can handle graph-structured data. DT-GBI constructs a decision tree while simultaneously constructing attributes for classification using GBI. In DT-GBI attributes, namely substructures useful for classification task, are constructed by GBI on the fly during the tree construction. We applied both GBI and DT-GBI to classification tasks of a real world hepatitis data. Three classification problems were solved in five experiments. In the first 4 experiments, DT-GBI was applied to build decision trees to classify 1) cirrhosis and non-cirrhosis (Experiments 1 and 2), 2) type C and type B (Experiment 3), and 3) positive and negative responses of interferon therapy (Experiment 4). As the patterns extracted in these experiments are thought discriminative, in the last experiment (Experiment 5) GBI was applied to extract descriptive patterns for interferon therapy. The preliminary results of experiments, both constructed decision trees and their predictive accuracies as well as extracted patterns, are reported in this paper. Some of the patterns match domain experts' experience and the overall results are encouraging. © Springer-Verlag Berlin Heidelberg 2007.

DOI： 10.1007/978-3-540-71009-7_2

Scopus

researchmap

記述言語：英語   掲載種別：論文集(書籍)内論文   出版者・発行元：John Wiley &amp; Sons, Inc.  

DOI： 10.1002/9780470073049.ch9

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/11731139_45

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd2006.html#NguyenOMMW06

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Institute of Electrical and Electronics Engineers Inc.  

An Auto-Regressive eXogenous input (ARX) model has been widely used in engineering fields to model dynamic response of a system to exogenous factors. A difficulty in this modeling is the determination of an appropriate model complexity, i.e., orders, for given data. In this paper, we develop a new and practical approach to determine the appropriate orders. Moreover, we apply the developed technique to a real marketing data, and analyze dynamic response character of sales amount to advertisement and sales promotion. In marketing study, static response of sales to some exogenous factors such as advertisement and sales promotion have been analyzed. However, if we can model dynamic response of sales to exogenous factors, more precise strategies of the sales to reduce the risk of the item stock management and increase the associated profit can be designed. © 2006 IEEE.

DOI： 10.1109/icdmw.2006.10

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer Verlag  

A graph mining method, Chunkingless Graph-Based Induction (Cl-GBI), finds typical patterns appearing in graph-structured data by the operation called chunkingless pairwise expansion, or pseudo-chunking which generates pseudo-nodes from selected pairs of nodes in the data. Cl-GBI enables to extract overlapping subgraphs, but it requires more time and space complexities than the older version GBI that employs real chunking. Thus, it happens that Cl-GBI cannot extract patterns that need be large enough to describe characteristics of data within a limited time and given computational resources. In such a case, extracted patterns maynot be so interesting for domain experts. To mine more discriminative patterns which cannot be extracted by the current Cl-GBI, we introduce a search algorithm in which patterns to be searched are guided by domain knowledge or interests of domain experts. We further experimentally show that the proposed method can efficiently extract more discriminative patterns using a real world dataset.

DOI： 10.1007/11961239_6

Scopus

researchmap

記述言語：日本語   出版者・発行元：The Society of Instrument and Control Engineers  

DOI： 10.11499/sicejl1962.44.307

CiNii Books

researchmap

その他リンク： https://jlc.jst.go.jp/DN/JALC/00250229404?from=CiNii

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Professional Book Center  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/ijcai/2005

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/BF03037635

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/11563983_22

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis2005.html#WashioAM05

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDM.2005.55

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2005.html#TermierRSOWM05

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDM.2005.100

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2005.html#WashioMM05

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/fuin/fuin66.html#InokuchiWM05

掲載種別：研究論文（学術雑誌）  

DOI： 10.1080/09528130512331315837

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

The volume of mass unsolicited electronic mail, often known as spam, has recently increased enormously and has become a serious threat to not only the Internet but also to society. A new spam detection method which uses document space density information has been proposed. Although the proposed method requires extensive e-mail traf-fic to acquire the necessary information, it can achieve perfect detection (i.e., both recall and precision is 100%) under practical conditions. This paper describes the memory management mechanism of this new spam detection method. Although the "Least Recently Used" strategy is the standard memory management strategy, we show that 1) the use of the direct-mapped cache can be used as a substitute for the LRU cache, and 2) "Retaining Multiply Accessed Entries" strategy can further improve the memory management performance and improve the theoretical recall rate for spam detection.

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/11564126_74

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pkdd/pkdd2005.html#WashioNM05

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer Verlag  

We have proposed a method called Decision Tree Graph-Based Induction (DT-GBI), which constructs a classifier (decision tree) for graph-structured data while simultaneously constructing attributes for classification. Graph-Based Induction (GBI) is utilized in DT-GBI for efficiently extracting typical patterns from graph-structured data by stepwise pair expansion (pairwise chunking). Attributes, i.e., substructures useful for classification task, are constructed by GBI on the fly while constructing a decision tree in DT-GBI. We applied DT-GBI to four classification tasks of hepatitis data using only the time-series data of blood inspection and urinalysis, which was provided by Chiba University Hospital. In the first and second experiments, the stages of fibrosis were used as classes and a decision tree was constructed for discriminating patients with F4 (cirrhosis) from patients with the other stages. In the third experiment, the types of hepatitis (B and C) were used as classes, and in the fourth experiment the effectiveness of interferon therapy was used as class label. The preliminary results of experiments, both constructed decision trees and their predictive accuracies, are reported in this paper. The validity of extracted patterns is now being evaluated by the domain experts (medical doctors). Some of the patterns match experts' experience and the overall results are encouraging. © Springer-Verlag Berlin Heidelberg 2005.

DOI： 10.1007/11423270_8

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

A machine learning technique called Graph-Based Induction (GBI) extracts patterns from graph structured data by stepwise pair expansion. GBI has been extended to 1) Beam-wise GBI(B-GBI) by incorporating a beam search to improve its search capability, and 2) Decision Tree Graph-Based Induction (DT-GBI) to construct a decision tree for graph-structured data. We applied B-GBI and DT-GBI to analyze the effectiveness of interferon therapy in the hepatitis dataset provided by Chiba University Hospital. Descriptive patterns were extracted by B-GBI and discriminative ones by DT-GBI using only the time sequence data of blood inspection and urinalysis. The discriminative patterns extracted by DT-GBI tend to be included in only relatively small number of patients and thus too specific. Thus, we tried to extract patterns which are both discriminative and descriptive by B-GBI. Furthermore, since there are exceptional situations (patients) with the extracted patterns, these patterns are further utilized to extract refined knowledge from the dataset. The preliminary results are reported with some of extracted patterns. © 2005 IEEE.

DOI： 10.1109/AMT.2005.1505269

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

Basket Analysis is mathematically characterized and extended to search families of sets in this paper. These theories indicate the possibility of various new approaches of data mining. We demonstrate the potential through proposal of a novel approach QARMINT. It performs complete mining of generic QARs within a low time complexity which has not been well addressed in the past work. Its performance evaluation shows high practicality. © 2005 IEEE.

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/11780496_46

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/jsai/jsai2005.html#WashioNMO05

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/11430919_74

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd2005.html#NguyenOMW05

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/ieicet/ieicet87d.html#YoshidaAWMHNFY04

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-540-30116-5_33

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pkdd/pkdd2004.html#NguyenWOM04

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Springer Verlag  

In this paper we discuss how graph mining system is applied to sales transaction data so as to understand consumer behavior. First, existing research of consumer behavior analysis for sequential purchase pattern is reviewed. Then we propose to represent the complicated customer purchase behavior by a directed graph retaining temporal information in a purchase sequence and apply a graph mining technique to analyze the frequent occurring patterns. In this paper we demonstrate through the case of healthy cooking oil analysis how graph mining technology helps us understand complex purchase behavior. © Springer-Verlag 2004.

DOI： 10.1007/978-3-540-30133-2_105

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：ACM  

DOI： 10.1145/1014052.1014107

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/kdd/kdd2004.html#YoshidaAWMHNFY04

掲載種別：研究論文（学術雑誌）  

researchmap

その他リンク： https://dblp.uni-trier.de/db/journals/ida/ida8.html#YoshidaWMW04

記述言語：日本語   出版者・発行元：The Society of Instrument and Control Engineers  

DOI： 10.11499/sicejl1962.42.480

CiNii Books

researchmap

その他リンク： https://jlc.jst.go.jp/DN/JALC/00229788444?from=CiNii

掲載種別：研究論文（学術雑誌）  

DOI： 10.1023/A:1021726221443

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/978-3-540-39644-4_12

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis2003.html#GeamsakulMYMW03

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-36175-8_6

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd2003.html#GeamsakulMYMW03

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer Verlag  

The needs of efficient and flexible information retrieval on multi-structural data stored in database and network are significantly growing. Most of the existing approaches is designed for a particular content and data structure, e.g., natural text and relational database. We propose a generic information retrieval method directly applicable to various types of contents and data structures. The power of this approach comes from the use of a generic and invariant feature information obtained from byte patterns in the files through some mathematical transformation. The experimental evaluation of the proposed approach for both artificial and real data indicates its high feasibility.

DOI： 10.1007/978-3-540-39592-8_69

Scopus

researchmap

記述言語：日本語   出版者・発行元：公益社団法人 日本化学会・情報化学部会  

.

DOI： 10.11546/cicsj.21.37

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

This article describes a watchdog program that discovers "meaningful" repair cases from a field service database. "Meaningful" cases are those judged worth probing further to prevent an epidemic of quality problems. Our system has employed that apriori algorithm, a data mining technique that efficiently performs the basket analysis. Our system proves that this data mining technique is not only useful in knowledge discovery but is also capable of performing the database watchdog task. The apriori algorithm automatically generals frequent itemsets from a large set of records. A frequent itemset is an arbitrary combination of values that appear more often than a threshold "minimum support". The algorithm often generates too many itemsets for quality engineers to review carefully in their daily work. Many itemsets do not provide sufficient information to investigate further. Hence, in order not to generate these valueless itemsets, the apriori algorithm is modified in two ways. One way is "basket analysis on objective and explanatory attributes" and the other is "itemset reduction." The advantage of our method is demonstrated with some experimental results.

DOI： 10.1002/eej.10034

Scopus

researchmap

記述言語：日本語   出版者・発行元：The Society of Instrument and Control Engineers  

DOI： 10.11499/sicejl1962.41.319

CiNii Books

researchmap

その他リンク： https://jlc.jst.go.jp/DN/JALC/00157320235?from=CiNii

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s101150200010

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-36182-0_44

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis2002.html#MatsudaMYW02

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45884-0_28

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dsp/dsp2002.html#HasegawaWI02

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45884-0_42

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dsp/dsp2002.html#WashioM02

掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/S1474-0346(02)00005-8

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/ICDM.2002.1183998

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icdm/icdm2002.html#YoshidaMW02

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45683-X_25

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pricai/pricai2002.html#WadaYMW02

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45683-X_26

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pricai/pricai2002.html#FujiwaraYMW02

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45683-X_29

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pricai/pricai2002.html#MatsudaMYW02

記述言語：日本語   出版者・発行元：The Japanese Society for Artificial Intelligence  

A machine learning technique called Graph-Based Induction (GBI) efficiently extracts typical patterns from graph data by stepwise pair expansion (pairwise chunking). In this paper, we introduce Graph-Based Induction for general graph structured data, which can handle directed/undirected, colored/uncolored graphs with/without (self) loop and with colored/uncolored links. We show that its time complexity is almost linear with the size of graph. We, further, show that GBI can effectively be applied to the extraction of typical patterns from DNA sequence data and organnochlorine compound data from which to generate classification rules, and that GBI also works as a feature construction component for other machine learning tools.

DOI： 10.1527/tjsai.16.363

CiNii Books

researchmap

その他リンク： https://jlc.jst.go.jp/DN/JALC/00151428081?from=CiNii

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/PL00011663

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer Verlag  

Basket Analysis is the most representative approach in recent study of data mining. However, it cannot be directly applied to the data including numeric attributes. In this paper, we propose an algorithm and performance measures for the selection and the discretization of numeric attributes in the data preprocessing stage for the wider application of Basket Analysis, and the performance is evaluated through the application to the meningitis data.

DOI： 10.1007/3-540-45548-5_72

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-44795-4_46

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/ecml/ecml2001.html#WashioMN01

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-44816-0_18

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/ida/ida2001.html#TerabeWM01

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45357-1_58

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd2001.html#WadaMW01

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer Verlag  

This paper describes automatic Web-page classification by using machine learning methods. Recently, the importance of portal site services is increasing including the search engine function on World Wide Web. Especially, the portal site such as Yahoo! service, which hierarchically classifies Web-pages into many categories, is becoming popular. However, the classification of Web-page into each category relies on man power, which costs much time and care. To alleviate this problem, we propose techniques to generate attributes by using co-occurrence analysis and to classify Web-page automatically based on machine learning. We apply these techniques to Web-pages on Yahoo! JAPAN and construct decision trees, which determine appropriate category for each Web-page. The performance of this proposed method is evaluated in terms of error rate, recall, and precision. The experimental evaluation demonstrates that this method provides acceptable accuracy with the classification of Web-page into top level categories on Yahoo! JAPAN.

DOI： 10.1007/3-540-45490-x_36

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Morgan Kaufmann  

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icml/icml2000.html#WashioMN00

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45571-X_49

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd2000.html#MatsudaHMW00

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-45372-5_2

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pkdd/pkdd2000.html#InokuchiWM00

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-44418-1_35

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis2000.html#HasegawaWIS00

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-44418-1_9

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis2000.html#MatsudaHMW00

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-46846-3_35

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis99.html#InokuchiWM98

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-48912-6_40

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd99.html#WadaHMW99

掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s101110050017

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-46846-3_33

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis99.html#HasegawaWI98

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-48912-6_20

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd99.html#TerabeKSWM99

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-48912-6_56

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd99.html#InokuchiWMKA99

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-46846-3_39

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis99.html#MatsudaHMWKA98

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Morgan Kaufmann  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/ijcai/99

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-64383-4_50

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pakdd/pakdd98.html#WashioM98

掲載種別：研究論文（学術雑誌）  

DOI： 10.1137/S1064827596305829

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：AAAI Press / The MIT Press  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/aaai/1998

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-49292-5_31

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/dis/dis98.html#WashioM98

掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/S0950-7051(98)00034-3

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Morgan Kaufmann  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/ijcai/1997

掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/S0004-3702(96)00060-4

researchmap

記述言語：日本語   出版者・発行元：Japan Ergonomics Society  

DOI： 10.5100/jje.32.Supplement_308

CiNii Books

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Springer  

DOI： 10.1007/3-540-61532-6_27

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pricai/pricai96.html#WashioM96

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Elsevier  

DOI： 10.1016/S0921-2647(06)80146-2

Scopus

researchmap

記述言語：日本語   出版者・発行元：日本知能情報ファジィ学会  

DOI： 10.3156/jfuzzy.5.5_958

researchmap

J-GLOBAL

researchmap

記述言語：日本語   出版者・発行元：応用物理学会有機分子・バイオエレクトロニクス分科会  

CiNii Books

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

&lt;p&gt;我々はSPoD-ExPAN超解像顕微鏡の画像推定に，Recursive Bayesian Computationを用いることを提案する。この推定手法は勾配法のような最適性評価関数の微分可能性を必要とせず、かつより大域的に最適な画像を得ることができると考えられる。本稿では、Recursive Bayesian Computation をカーネル密度推定を用いて実装する。また、推定に正則化を導入し、その推定精度を正則化を行わない場合と比較検証する。&lt;/p&gt;

researchmap

J-GLOBAL

researchmap

記述言語：日本語   出版者・発行元：応用物理学会有機分子・バイオエレクトロニクス分科会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：化学同人  

CiNii Books

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

&lt;p&gt;近年の超解像蛍光顕微鏡技術開発により、電子顕微鏡に頼らずに細胞内の微細構造の可視化が可能になった。しかし、細胞挙動の理解を今後進めるには、細胞内の構造のみならず、情報伝達物質の動き、温度、酵素反応等の生理パラメータを超解像イメージングで計測し、そこから潜在的要因の検出や状態の推定をすることが必要になってくる。本論文では、蛍光プローブ開発と生理機能超解像イメージングへ向けた取り組みについて発表する。&lt;/p&gt;

researchmap

記述言語：英語   出版者・発行元：Springer Verlag  

Scopus

researchmap

記述言語：日本語   出版者・発行元：人工知能学会 ; 2014-  

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1004/00001792/

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：システム制御情報学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人日本物理学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

直接方策探索法(Direct Policy Search:DPS)に基づく強化学習は,大規模,かつ複雑なマルコフ決定過程に対する解法として近年,注目を集めている.DPSはこれまで様々な実問題に応用され,目覚しい成果を挙げているが,方策モデルの選択法に関してはこれまでほとんど議論されてこなかった.本研究では,方策を重み付き最尤推定法により学習する新しい手法,weighted likelihood policy search(WLPS)法を提案する.WLPSはDPSの統計推論問題として自然な解釈を与え,統計学習分野で確立されているモデル選択に関する方法論を直接,DPSに適用することを可能にする.本研究では,情報量基準の考えを方策探索に応用し,重み付き尤度を規範とした方策モデルの選択法を提案する.

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

異常箇所同定はシステムの故障診断などにおいて重要な問題である.本稿ではグラフィカルモデルに基づく異常箇所同定のための新たなグラフィカルモデル推定法を提案する.提案法では精度行列の各行・列の変化に制約を加える正則化項を導入し,Alternating Direction Methodを用いてアルゴリズムを構築した.また,実データを用いた実験において,提案法により異常箇所同定の精度が向上することを確認した.

CiNii Books

researchmap

記述言語：日本語  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1004/00007980/

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：英語   出版者・発行元：人工知能学会  

DOI： 10.11517/pjsai.jsai2012.0_4b1r24

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：東京 : 人工知能学会  

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I023433809

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

本稿では,2つの劣モジュラ関数の差で表される評価関数の最小化のための厳密解法を提案する.提案するアルゴリズムは,分枝限定法の枠組みの中で,本問題の構造の特殊性を利用したものである.任意の集合関数最適化はこの問題へと定式化する事が可能であり,機械学習で扱われる多くの問題においても広く適用可能である.本稿では,特に特徴選択へ適用した場合の数値例を示す.

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

我々は先行研究として,すべての観測変数が二値を取るデータセットからそのデータ生成過程を一意に同定する推定法を提案した.しかし,この先行研究において用いられる推定指標は統計的に最適な推定を保証するものではない.本稿では,この点について統計的により適切な推定指標を提案する.この指標は観測データから作られる分割表の独立性を直接評価するものであり,先行研究に比して精度の向上が期待できる.実験により,我々はこの推定指標がより良い性能を示すことを確認した.

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

グラフィカル・モデルの構造推定は確率変数間の依存関係を解明する上で重要な問題である.近年,スパース正則化を用いて高次元データに内在する依存関係を推定する手法の研究が活発になされている.本稿ではクラスタ構造を有するグラフィカル・ガウシアン・モデル(GGM)の構造推定手法を提案する.我々は従来のGGM推定手法(l_1正則化最尤法)がグラフ・ラプラシアンのトレースノルム正則化として解釈できることを示し,それに基づいてクラスタ構造を考慮した新たな正則化手法を提案する.また,数値実験において提案法の有効性を確認する.

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：東京 : 電子情報通信学会  

CiNii Books

researchmap

その他リンク： https://ndlsearch.ndl.go.jp/books/R000000004-I023346021

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

「情報爆発」によって生み出されるデータを,有益な知識源として利用しようというニーズは大きい.しかし,その膨大さゆえにそのままでは有効活用が困難である場合が多い.そこで,膨大なデータを分析し,利用可能な知識を抽出するデータマイニング技術の開発が必要不可欠となっている.特に,多数の項目で表される事例の集まりである高次元データを対象とした場合,次元やデータの増加に対してスケーラブルな処理が問題となるのみならず,高次元データに特有な分析精度や効率を低下させる次元の呪いと呼ばれる効果が知られている.本稿ではこの問題とその対応技術ついて概説する.

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

確率変数間の依存構造はデータの背後にある生成過程と密接に関連している.そのため,変数間の依存構造の推定はデータマイニングにおける重要な問題となっている.特に,複数の異なる環境下で収集されたデータが一部の変数間に共通した依存構造を持つ場合,背後に環境条件によらないより本質的な依存関係が存在する可能性が示唆される.本研究では,グラフィカル・ガウシアン・モデル(GGM)について複数のGGM間で共有される部分構造を推定する手法を提案する.提案法は従来のGGMの構造推定手法の自然な拡張となっており,ブロック座標降下法により効率的に解を得ることができる.数値実験により,提案法が従来法と同等のGGM構造推定精度を有しつつ,高い精度で共通部分構造を推定できることを確認した.

CiNii Books

researchmap

J-GLOBAL

researchmap

記述言語：英語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：英語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

近年、大量の観測データからそのデータがどのような生成過程によって生成されたかを知りたいというニーズが高まっている。そして最近、データが連続変数の場合はデータの非正規性を利用することで生成過程を同定できる場合があることがわかってきた。本発表ではデータが二値変数の場合における新しいデータ生成過程のモデルBSAM (Boolean Skew Acyclic Model)とその推定アルゴリズムを提案する。

DOI： 10.11517/pjsai.jsai2011.0_2e36

CiNii Books

J-GLOBAL

researchmap

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

対象に関する多数の変数や属性を容易に同時測定することができる時代となり,どの変数が他のどの変数に影響してデータが生成されたか,あるいは要素間のどのような関係が次にどのような関係を生んだか,即ちデータ生成過程を体系的に把握して対象を理解したい,というニーズが高まっている.長い間,この目的にはグラフィカルモデリングや時系列データ解析など技術が使われてきたが,近年,新たな技術展開が見られている.我々の研究室ではグラフマイニングや統計的因果推論,最適化アルゴリズムなどの研究を通じて,多数の変数や要素間のデータ生成因果の解析に基づき,データ生成過程やそこに埋め込まれた複雑な関係を理解するリレーショナルデータマイニング手法の開発に取り組んでいる.ここでは,これらの研究の取り組みと具体的応用解析例を紹介し,データ生成過程の理解を目標とするリレーショナルデータマイニングの可能性を展望する.

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

高次元確率空間においては、次元の呪いにより有限個数データの標本分布がその母集団分布を的確に反映しない。また、同じく確率密度関数によっては、確率空間内の非常に限られた範囲でしか有意な大きさをとらない場合が多い。本報では、これらの効果により、期待値のベイズ推定が大きなバイアスを持つことを示し、どのような場合に機械学習やデータマイニングに大きな影響を与えるかを理論的考察と実験によって明らかにする。

DOI： 10.11517/pjsai.jsai2010.0_1a14

CiNii Books

J-GLOBAL

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

本研究では，データの非ガウス性を利用する独立成分分析を用いて，様々な外的刺激に対する遺伝子発現データについて，それぞれ外性的に発現している遺伝子の同定を行い考察を行う．

DOI： 10.11517/pjsai.jsai2009.0_2c13

CiNii Books

J-GLOBAL

researchmap

記述言語：英語   出版者・発行元：SPRINGER HEIDELBERG  

DOI： 10.1007/s10463-008-0208-y

Web of Science

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1004/00006982/

記述言語：日本語  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1004/00006719/

記述言語：日本語  

CiNii Books

researchmap

記述言語：日本語  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人 人工知能学会  

本研究ではバイオインフォマティクスのデータから飽和頻出木をマイニングすることを試みる。対象データの構造はDAGなので、我々のツリーマイニングアルゴリズムDryadeをＤＡＧに適用可能なように改良した。実験でこの効果を確かめる。

DOI： 10.11517/pjsai.JSAI06.0.7.0

researchmap

記述言語：日本語   出版者・発行元：一般社団法人情報処理学会  

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1004/00006019/

記述言語：日本語   出版者・発行元：一般社団法人電子情報通信学会  

データマイニングは,最近急速にクローズアップされてきている技術である.しかし,それは特定の決まった手法を指すのではなく,データ解析を行う様々な要素技術の総体である.このことがデータマイニングの概念を分かり難くし,パターン認識のような他の分野との技術的融合の障害になっていると思われる。この論説では,データマイニング研究の現状や最近の動向,課題を概観し,それをパターン認識技術と融合する際に,双方の分野の概念の相違から来る問題について考察する.

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：Human Interface Society  

Current difficulty to formulate complex systems may be resolved by introducing scale constraints thanks to the development of advanced information technology. This paper concerns the identification of the first principle equation that governs complex system beviour by applying machine learning and trial study of the relationship between the public affinity and earthquake risk. The basic research of the discovery of the first principle equation has been made in the area of artificial intelligence, a kind of engineering. This outcome was extended to socio-psychology. Thus the framework is considered to be a symbiosis of natural and cultural sciences.

DOI： 10.11184/hisrm.4.2_1

CiNii Books

researchmap

その他リンク： https://jlc.jst.go.jp/DN/JALC/10015804463?from=CiNii

記述言語：日本語   出版者・発行元：一般社団法人情報処理学会  

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1001/00063806/

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1004/00005190/

記述言語：日本語   出版者・発行元：一般社団法人 日本機械学会  

Since design knowledge plays a crucial role in design process, we are conducting research on design reasoning model that takes notice of design knowledge. To make a design reasoning model, we have already proposed a reasoning framework of design and analyzed basic operations for the framework, and then propose a synthesis language for describing design operational knowledge. In this paper, we propose our implementation of this model, and verify it to compare design processes replayed in the system with ones of actual design sessions.

DOI： 10.1299/jsmedsd.2001.10.285

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：一般社団法人 日本機械学会  

Since design knowledge plays a crucial role in design process, we are conducting research on design reasoning model that takes notice of design knowledge. To make a design reasoning model, we have already proposed a reasoning framework of design and analyzed basic operations for the framework. However, since these operations are just fragments of whole design process, it is necessary to describe design operational knowledge that can control the entire reasoning operations by constructing the sequence of these fragments. So, in this paper, we propose a synthesis language for describing design operational knowledge.

DOI： 10.1299/jsmedsd.2001.10.281

CiNii Books

researchmap

記述言語：日本語   出版者・発行元：人工知能学会  

CiNii Books

researchmap

記述言語：日本語  

CiNii Books

researchmap

記述言語：日本語  

This paper proposes methods to derive the exogenously-driven causality of a physical system based on the intrinsic assumptions of evolutional causality and functional causality of each of the system's internal physical laws. The exogenously-driven causality of a physical system has previously been considered as being a feature of the entire system. The work reported here demonstrates that a major portion of the knowledge can be deriven from the intrinsic causal assumptions associated with each elementary physical law in the system description without requiring any information on external mechanisms. First, heuristics to extract the system-independent causal assumptions of each physical law are proposed. Second, a knowledge representation that describes the quantitative and causal assumptions of each physical law is established. This representation consists of 'assumptive structural equations'. Third, a method to derive candidate exogenous variables, causal structures, and mythical causality of the system is demonstrated based on the assumptive structural equations. Fourth, algebraic rules to derive various equations involving the knowledge of exogenously-driven causality of the system are shown. This methodology should be of use in simulation, design planning, and diagnosis of physical systems.

CiNii Books

researchmap