担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.3390/app11209728

researchmap

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1109/ACCESS.2021.3073945

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：IOS Press  

DOI： 10.3233/ida-184283

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 BV  

DOI： 10.1016/j.knosys.2019.05.012

researchmap

担当区分：最終著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

researchmap

担当区分：最終著者   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：ACM  

DOI： 10.1145/3205651.3205675

researchmap

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s13755-017-0031-z

researchmap

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.artmed.2017.06.012

Web of Science

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：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.  

In data mining and machine learning, the definition of the distance between two data points substantially affects clustering and classification tasks. We propose a distance metric learning (DML) method for multi-label clustering, that uses evolutionary multi-objective optimization and a cluster validity measure with a neighbor relation that simultaneously evaluates inter-and intra-clusters. The proposed method produces clustering results considering multiple class labels and allows the induction of knowledge regarding relations between class labels in multi-label clustering or between objective functions and elements in transform matrix. Experimental results have shown that the proposed DML method produces better transform matrices than single-objective optimization and is helpful in finding the attributes that affect the trade-off relationship among objective functions.

DOI： 10.1109/CEC.2015.7257255

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  

We propose a pattern mining algorithm for numerical multidimensional event sequences, called cluster sequence mining (CSM). CSM extracts patterns with a pair of clusters that satisfies space proximity of the individual clusters and time proximity in time intervals between events from different clusters. CSM is an extension of a unique algorithm (co-occurrence cluster mining (CCM)), considering the order of events and the distribution of time intervals. The probability density of the time intervals is inferred by utilizing Bayesian inference for robustness against uncertainty. In an experiment using synthetic data, we confirmed that CSM is capable of extracting clusters with high F-measure and low estimation error of the time interval distribution even under uncertainty. CSM was applied to an earthquake event sequence in Japan after the 2011 Tohoku Earthquake to infer causality of earthquake occurrences. The results demonstrate that CSM suggests some high affecting/affected areas in the subduction zone farther away from the main shock of the Tohoku Earthquake.

DOI： 10.1007/978-3-319-18032-8_23

Scopus

researchmap

担当区分：最終著者   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）  

researchmap

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Institute of Electronics, Information and Communications Engineers (IEICE)  

DOI： 10.1587/transinf.2023edl8084

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 BV  

DOI： 10.1016/j.renene.2024.120243

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.48550/arXiv.2404.16881

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.48550/arXiv.2404.10299

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：American Meteorological Society  

Abstract

Surrogate modeling is one of the most promising applications of deep learning techniques in meteorology. The purpose of this study was to downscale surface wind fields in a gridded format at a much lower computational load. We employed a superresolution convolutional neural network (SRCNN) as a surrogate model and created a 20-member ensemble by training the same SRCNN model with different random seeds. The downscaling accuracy of the ensemble mean remained stable throughout a year and was consistently better than that of the input wind fields. It was confirmed that 1) the ensemble spread was efficiently created and that 2) the ensemble mean was superior to individual ensemble members and 3) robust to the presence of outlier members. Training, validation, and test data for 10 years were computed via our nested mesoscale weather forecast models not derived from public analysis datasets or real observations. The predictands were 1-km gridded surface zonal and meridional winds, of which the domain was defined as a 180 km × 180 km area around Tokyo, Japan. The predictors included 5-km gridded surface zonal and meridional winds, temperature, humidity, vertical gradient of the potential temperature, elevation, and land-to-water ratio as well as 1-km gridded elevation and land-to-water ratio. Although a perfect surrogate of the weather forecast model could not be achieved, the SRCNN downscaling accuracy could likely enable us to apply this approach in high-resolution advection simulations, considering its overwhelmingly high prediction speed.

DOI： 10.1175/aies-d-23-0007.1

researchmap

DOI： 10.1016/j.jpowsour.2023.233059

researchmap

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.48550/arXiv.2308.10283

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：MDPI AG  

Deep learning using an end-to-end convolutional neural network (ConvNet) has been applied to several electroencephalography (EEG)-based brain–computer interface tasks to extract feature maps and classify the target output. However, the EEG analysis remains challenging since it requires consideration of various architectural design components that influence the representational ability of extracted features. This study proposes an EEG-based emotion classification model called the multi-kernel temporal and spatial convolution network (MultiT-S ConvNet). The multi-scale kernel is used in the model to learn various time resolutions, and separable convolutions are applied to find related spatial patterns. In addition, we enhanced both the temporal and spatial filters with a lightweight gating mechanism. To validate the performance and classification accuracy of MultiT-S ConvNet, we conduct subject-dependent and subject-independent experiments on EEG-based emotion datasets: DEAP and SEED. Compared with existing methods, MultiT-S ConvNet outperforms with higher accuracy results and a few trainable parameters. Moreover, the proposed multi-scale module in temporal filtering enables extracting a wide range of EEG representations, covering short- to long-wavelength components. This module could be further implemented in any model of EEG-based convolution networks, and its ability potentially improves the model’s learning capacity.

DOI： 10.3390/s22218250

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：Japan Prosthodontic Society  

DOI： 10.2186/jpr.jpr_d_20_00260

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：MDPI AG  

Numerical modeling for geothermal reservoir engineering is a crucial process to evaluate the performance of the reservoir and to develop strategies for the future development. The governing equations in the geothermal reservoir models consist of several constitutive parameters, and each parameter is given to a large number of simulation grids. Thus, the combinations of parameters we need to estimate are almost limitless. Although several inverse analysis algorithms have been developed, determining the constitutive parameters in the reservoir model is still a matter of trial-and-error estimation in actual practice, and is largely based on the experience of the analyst. There are several parameters which control the hydrothermal processes in the geothermal reservoir modeling. In this study, as an initial challenge, we focus on permeability, which is one of the most important parameters for the modeling. We propose a machine-learning-based method to estimate permeability distributions using measurable data. A large number of learning data were prepared by a geothermal reservoir simulator capable of calculating pressure and temperature distributions in the natural state with different permeability distributions. Several machine learning algorithms (i.e., linear regression, ridge regression, Lasso regression, support vector regression (SVR), multilayer perceptron (MLP), random forest, gradient boosting, and the k-nearest neighbor algorithm) were applied to learn the relationship between the permeability and the pressure and temperature distributions. By comparing the feature importance and the scores of estimations, random forest using pressure differences as feature variables provided the best estimation (the training score of 0.979 and the test score of 0.789). Since it was learned independently of the grids and locations, this model is expected to be generalized. It was also found that estimation is possible to some extent, even for different heat source conditions. This study is a successful demonstration of the first step in achieving the goal of new data-driven geothermal reservoir engineering, which will be developed and enhanced with the knowledge of information science.

DOI： 10.3390/geosciences12030130

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

STUDY OBJECTIVES: The present study investigated the hypothesis that subjects with primary sleep bruxism (SB) exhibit masseter and cortical hyperactivities during quiet sleep periods that are associated with a high frequency of rhythmic masticatory muscle activity (RMMA). METHODS: Fifteen SB and ten control participants underwent polysomnographic recordings. The frequencies of oromotor events and arousals and the percentage of arousals with oromotor events were assessed. Masseter muscle tone during sleep was quantified using a cluster analysis. Electroencephalography power and heart rate variability were quantified and then compared between the two groups and among sleep stages. RESULTS: The frequency of RMMA and percentage of arousals with RMMA were significantly higher in SB subjects than in controls in all stages, while these variables for non-rhythmic oromotor events did not significantly differ between the groups. In SB subjects, the frequency of RMMA was the highest in stage N1 and the lowest in stages N3 and R, while the percentage of arousals with RMMA was higher in stage N3 than stages N1 and R. The cluster analysis classified masseter activity during sleep into two clusters for masseter tone and contractions. Masseter muscle tone showed typical stage-dependent changes in both groups, but did not significantly differ between the groups. Furthermore, no significant differences were observed in electroencephalography power or heart rate variability between the groups. CONCLUSION: Young SB subjects exhibited sleep stage-dependent increases in the responsiveness of RMMA to transient arousals, but did not show masseter or cortical hyperactivity during sleep.

DOI： 10.1093/sleep/zsab207

PubMed

researchmap

担当区分：最終著者,　責任著者   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/IJCNN52387.2021.9533606

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/ijcnn/ijcnn2021.html#ThanasutivesNF21

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/ijcnn52387.2021.9533310

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

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：MDPI AG  

Two of the biggest challenges in building models for detecting emotions from electroencephalography (EEG) devices are the relatively small amount of labeled samples and the strong variability of signal feature distributions between different subjects. In this study, we propose a context-generalized model that tackles the data constraints and subject variability simultaneously using a deep neural network architecture optimized for normally distributed subject-independent feature embeddings. Variational autoencoders (VAEs) at the input level allow the lower feature layers of the model to be trained on both labeled and unlabeled samples, maximizing the use of the limited data resources. Meanwhile, variational regularization encourages the model to learn Gaussian-distributed feature embeddings, resulting in robustness to small dataset imbalances. Subject-adversarial regularization applied to the bi-lateral features further enforces subject-independence on the final feature embedding used for emotion classification. The results from subject-independent performance experiments on the SEED and DEAP EEG-emotion datasets show that our model generalizes better across subjects than other state-of-the-art feature embeddings when paired with deep learning classifiers. Furthermore, qualitative analysis of the embedding space reveals that our proposed subject-invariant bi-lateral variational domain adversarial neural network (BiVDANN) architecture may improve the subject-independent performance by discovering normally distributed features.

DOI： 10.3390/s21051792

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）  

DOI： 10.5220/0010200701500157

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）  

researchmap

担当区分：最終著者,　責任著者   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）  

researchmap

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

担当区分：最終著者,　責任著者   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）  

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）  

DOI： 10.1109/smc42975.2020.9283107

researchmap

担当区分：最終著者,　責任著者   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： 10.1016/j.procs.2020.09.085

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）  

DOI： 10.1007/978-3-030-59277-6_28

researchmap

担当区分：最終著者,　責任著者   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）  

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）  

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  

This is an extension from a selected paper from JSAI2019. With the proliferation of wearable devices and the inflow of new health data, artificial intelligence is expected to revolutionize the field of wellness and health management by providing potential tools for analyzing harmful conditions like prolonged stress. Currently, one of the standard measurements used by medical practitioners to measure stress is heart rate variability (HRV), a set of numerical indices that reflect autonomic balance. However, recent advances in machine learning have shown that learned features tend to outperform hand-crafted features. In this work we propose a more expressive intermediate data representation based on Lomb-Scargle periodograms combined with the feature learning capabilities of deep learning. Using stress data from naturalistic work activities, we tested different shallow and deep learning architectures and show that significant improvements can be achieved compared to traditional HRV indices. Results show that models trained on our spectral-temporal representation significantly outperform models trained on traditional HRV indices for predicting naturalistic work stress.

DOI： 10.1007/978-3-030-39878-1_24

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  

This is an extension from a selected paper from JSAI2019. Recent studies in affective computing have facilitated and stimulated the development of systems and sensors that can recognize and interpret human affects. Affective computing has been applied in various domains, and one of the applied domains is in the marketing area to increase the consumers’ appeal and attraction. In particular, advertisements (ads) can convey amounts of information in a short time. Therefore, using physiological responses can help to acquire a user’s feedback and obtain an advantage. This study proposes non-invasive affect recognition in each scene of an advertising video using electroencephalogram (EEG), electrocardiogram (ECG) and eye-tracking. The preliminary analysis of EEG shows the relationship between scene feeling score and emotional affects regarding physiological responses. Hence, we also trained two types of recognition models: window recognition and sequence learning. The models learned from the physiological responses and questionnaires on a user’s preference in each ad scene.

DOI： 10.1007/978-3-030-39878-1_3

Scopus

researchmap

担当区分：筆頭著者,　最終著者   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）  

researchmap

担当区分：最終著者   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）  

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）  

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）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語  

researchmap

担当区分：最終著者   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  

Data stream classifiers that can withstand unusual phenomena in an evolving data stream, such as concept drift and concept evolution, are highly desirable for data stream mining. Most existing methods deal with such phenomena in a supervised manner, which is costly in a real-world scenario. To address this shortcoming, we propose a concept drift detection approach that combines our approach with a semi-supervised adaptive incremental neural gas (A2ING) classifier. Our approach makes use of A2ING's graph topology structure to detect changes in a data stream. We derive a graph's instability around its decision boundary and find the difference in prior and posterior distributions of the criteria. The empirical results show the effectiveness of our method. The classifier requires a relatively low number of true labels compared to existing approaches and shows high effectiveness in change detection.

DOI： 10.1109/ICTAI.2017.00077

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：Springer  

DOI： 10.1007/978-3-319-78301-7_3

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：Institute of Electrical and Electronics Engineers Inc.  

Conventional feature selection methods may not provide sufficient means to deal with the diverse growing amount of data nowadays. Evolutionary Distance Metric Learning (EDML) relies on an evolutionary approach in its distance metric learning process
this process in case of diagonal EDML can be viewed as an embedded feature weighting one. However, such process is done simultaneously on all features and does not explicitly select the features. This paper introduces a new hybrid system R-EDML, combining the sequential decision making of Reinforcement Learning (RL) with the evolutionary feature prioritizing process of EDML in clustering. The goal is to create a feature selection control strategy that aims to optimize the input space by reducing the number of selected features while maintaining the clustering performance. This can lead to future data collection time and cost reduction. In the proposed method, features represented by the elements of EDML distance transformation matrices are prioritized by a differential evolution algorithm. Then a selection control strategy using reinforcement learning is learned by sequentially inserting and evaluating the prioritized elements. This process is repeated with the aim to optimize the matrices by filtering the elements used in them. The outcome is the selection of the best R-EDML generation matrices with the least number of elements possible. R-EDML was compared to normal EDML in terms of feature selection and accuracy. Results show a decrease in the number of features compared to EDML, while maintaining a similar accuracy level.

DOI： 10.1109/SSCI.2017.8280866

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）  

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：Institute of Electrical and Electronics Engineers Inc.  

This paper presents a framework for adaptive multimodal emotion recognition based on signal stability as a context. To verify the efficacy of the method, experiments were conducted using a dataset of brainwave and physiological signals (EEG, ECG, GSR) captured from nine subjects listening to music. The proposed method uses a combination of signal-based features as well as accelerometer data to quantify the approximate reliability of each modality. In contrast to existing approaches, unstable modalities are not rejected outright, instead their relative contribution is dynamically adapted based on a corresponding stability index. In the case of EEG, the stability index was calculated using an artifact rejection technique, while for the ECG and GSR modalities it was calculated based on body movement detected through accelerometers. The experimental results show that temporally varying the relative contribution of each modality can improve emotion recognition performance.

DOI： 10.1109/ACIIW.2017.8272584

Scopus

researchmap

担当区分：最終著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1527/tjsai.D-SGAI02

Scopus

researchmap

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1527/tjsai.G-SGAI05

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）  

researchmap

担当区分：最終著者   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）  

© 2017 IEEE. ART-2a has been shown to be effective against stream data clustering with unknown number of cluster in nature. As data grows, ART-2a running time become a major problem. We proposed a new algorithm, ART-2b, whose runtime performance is linear to the number of input instances, while still maintaining similar clustering result to ART-2a.

DOI： 10.1109/BigData.2017.8258551

Scopus

researchmap

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.11238/mammalianscience.57.19

researchmap

担当区分：最終著者   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）  

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）  

researchmap

担当区分：最終著者   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）  

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）  

researchmap

記述言語：英語  

researchmap

担当区分：最終著者   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： 10.1007/978-3-319-49049-6_12

Web of Science

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.52731/iee.v2.i3.85

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）  

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）  

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 BV  

DOI： 10.1016/j.renene.2016.02.051

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  

In the construction of computational models to recognize emotional state, emotion reporting continuously in time is essential based on the assumption that emotional responses of a human to certain stimuli could vary over time. However, currently existing methods to annotate emotion in temporal continuous fashion are confronting various types of challenges. Therefore, the manipulation of the annotated emotion prior to labeling training samples is necessary. In this work, we present an early attempt to manipulate the emotion annotated in arousal-valence space by applying three different signal filtering techniques to smooth annotation data
moving average filter, Savitzky-Golay filter, and median filter. We conducted experiments of emotion recognition in music listening tasks employing brainwave signals recorded from an electroencephalogram (EEG). Smoothed annotation data were used to label the features extracted from EEG signals to train emotion recognizers using classification and regression techniques. Our empirical results indicated the potential of the moving average filter that could increase the performance of emotion recognition evaluated in subject-independent fashion.

DOI： 10.1007/978-3-319-60675-0_10

Scopus

researchmap

記述言語：英語  

researchmap

担当区分：最終著者   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/IIAI-AAI.2016.28

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/iiaiaai/iiaiaai2016.html#KashimaMHF16

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Maruzen Co., Ltd.  

Research on emotion recognition using electroencephalogram (EEG) of subjects listening to music has become more active in the past decade. However, previous works did not consider emotional oscillations within a single musical piece. In this research, we propose a continuous music-emotion recognition approach based on brainwave signals. While considering the subject-dependent and changing-over-time characteristics of emotion, our experiment included self-reporting and continuous emotion annotation in the arousal-valence space. Fractal dimension (FD) and power spectral density (PSD) approaches were adopted to extract informative features from raw EEG signals and then we applied emotion classification algorithms to discriminate binary classes of emotion. According to our experimental results, FD slightly outperformed PSD approach both in arousal and valence classification, and FD was found to have the higher correlation with emotion reports than PSD. In addition, continuous emotion recognition during music listening based on EEG was found to be an effective method for tracking emotional reporting oscillations and provides an opportunity to better understand human emotional processes.

DOI： 10.1587/transinf.2015EDP7251

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

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  

To gain a better understanding of how neural ensembles communicate and process information, neural decoding algorithms are used to extract information encoded in their spiking activity. Bayesian decoding is one of the most used neural population decoding approaches to extract information from the ensemble spiking activity of rat hippocampal neurons. Recently it has been shown how Bayesian decoding can be implemented without the intermediate step of sorting spike waveforms into groups of single units. Here we extend the approach in order to make it suitable for online encoding/decoding scenarios that require real-time decoding such as brain-machine interfaces. We propose an online algorithm for the Bayesian decoding that reduces the time required for decoding neural populations, resulting in a real-time capable decoding framework. More specifically, we improve the speed of the probability density estimation step, which is the most essential and the most expensive computation of the spike-sorting-less decoding process, by developing a kernel density compression algorithm. In contrary to existing online kernel compression techniques, rather than optimizing for the minimum estimation error caused by kernels compression, the proposed method compresses kernels on the basis of the distance between the merging component and its most similar neighbor. Thus, without costly optimization, the proposed method has very low compression latency with a small and manageable estimation error. In addition, the proposed bandwidth matching method for Gaussian kernels merging has an interesting mathematical property whereby optimization in the estimation of the probability density function can be performed efficiently, resulting in a faster decoding speed. We successfully applied the proposed kernel compression algorithm to the Bayesian decoding framework to reconstruct positions of a freely moving rat from hippocampal unsorted spikes, with significant improvements in the decoding speed and acceptable decoding error.

DOI： 10.1016/j.knosys.2015.09.013

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：SCITEPRESS  

DOI： 10.5220/0005711000470055

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）   出版者・発行元：World Scientific  

DOI： 10.1142/9789814730464_0008

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）   出版者・発行元：World Scientific  

DOI： 10.1142/9789814730464_0004

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）   出版者・発行元：World Scientific  

DOI： 10.1142/9789814730464_0005

researchmap

担当区分：最終著者   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 quality of a good sleep is important for a healthy life. Recently, several sleep analysis products have emerged on the market
however, many of them require additional hardware or there is a lack of scientific evidence regarding their clinical efficacy. This paper proposes a novel method for discovering the sleep pattern via clustering of sound events. The sleep-related sound clips are extracted from sound recordings obtained when sleeping. Then, various self-organizing map algorithms are applied to the extracted sound data. We demonstrate the superiority of Kullback-Leibler divergence and obtain the cluster maps to visualize the distribution and changing patterns of sleep-related events during the sleep. Also, we perform a comparative interpretation between sleep stage sequences and obtained cluster maps. The proposed method requires few additional hardware, and its consistency with the medical evidence proves its reliability.

DOI： 10.1007/978-3-319-42007-3_40

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）  

DOI： 10.1007/978-3-319-25032-8_23

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  

Automated mental health analysis of stress could lead towards diagnosis tools that can be used in environments such as clinics, schools and corporations. However, attempts at building general models are often limited by the subjectivity of physiological stress responses. This work aims to discover the effects of combining data from physiological signals and psychological context from work activities when building a machine-learned model of mental stress. A software application was built to guide subjects through a monitoring process which allowed pre and post-assessment of psychological context through various stress-related annotation modules including the Cohen Stress Scale and the COPE inventory. Meanwhile, wearable sensors tracked physiological data in the form of heart beats. Tests were performed on this data by building supervised and unsupervised machine-learned models. Results show a general increase in classification performance when psychological context data is integrated into the models. Furthermore, models present similar performance using either questionnaire answers or coping profile scores.

DOI： 10.1007/978-3-319-46218-9_9

Scopus

researchmap

記述言語：英語  

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）  

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）  

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）  

researchmap

記述言語：英語  

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：JMIR Publications  

DOI： 10.2196/medinform.3803

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  

The predictive technique proposed in this project was initially designed for an indoor smart environment wherein intrusive tracking techniques, such as cameras, mobile phones, and GPS tracking systems, could not be appropriately utilized. Instead, we installed simple motion detection sensors in various areas of the experimental space and observed movements. However, the data collected cannot provide as much information about human mobility as data from a GPS or mobile phone. In this paper, we conducted an exhaustive analysis to determine the predictability of future mobility of people using only this limited dataset. Furthermore, we proposed an aperiodic and periodic predictive technique for long-term human mobility prediction that works well with our limited dataset. The evaluation of the dataset collected of the movement and daily activity in the smart space for three months shows that our model is able to predict future mobility and activities of participants in the smart environment setting with high accuracy – even for a month in advance.

DOI： 10.1007/978-3-319-14723-9_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）   出版者・発行元：WORLD SCIENTIFIC PUBL CO PTE LTD  

The emergence of e-patient has encouraged people to be more literate about healthcare subject. However, health information seeking remains problematic for most non-medical professionals (consumers). Most consumers are not knowledgeable with medical terminology. The diversity of consumer's familiarity with health domain also leads to frustration since the information presented may fall outside the consumer's comprehension level. This research aims to assist consumers obtain understandable health information by designing adaptive personalization approach in health information retrieval system. The proposed approach constructs user model dynamically based on the interaction with search engine. The user model captures contextual attributes and the familiarity level with health topic in a weighted semantic network. A node represents a topic of interest and its familiarity level and a weighted-link shows semantic similarity between nodes, which refers to Unified Medical Language System Semantic Network.

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：Springer Verlag  

The trend of multimodal interaction in interactive gaming has grown significantly as demonstrated for example by the wide acceptance of the Wii Remote and the Kinect as tools not just for commercial games but for game research as well. Furthermore, using the player’s affective state as an additional input for game manipulation has opened the realm of affective gaming. In this paper, we analyzed the affective states of players prior to and after witnessing a scary event in a survival horror game. Player affect data were collected through our own affect annotation tool that allows the player to report his affect labels while watching his recorded gameplay and facial expressions. The affect data were then used for training prediction models with the player’s brainwave and heart rate signals, as well as keyboard–mouse activities collected during gameplay. Our results show that (i) players are likely to get more fearful of a scary event when they are in the suspense state and that (ii) heart rate is a good candidate for detecting player affect. Using our results, game designers can maximize the fear level of the player by slowly building tension until the suspense state and showing a scary event after that. We believe that this approach can be applied to the analyses of different sets of emotions in other games as well.

DOI： 10.1007/s12193-014-0153-4

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語  

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）  

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 INT PUBLISHING AG  

Argo float is a small and light-weight drifting buoy to measure oceanic temperature and salinity. More than 3,600 floats are always working for globally-covered ocean monitoring, and the accumulated big ocean observation data helps many studies such as investigation into climate change mechanism. However, the observed temperature and salinity data sometimes involves errors. Since automatic detection and correction of the errors is difficult due to ununiform observation reliability and the necessity of specifying error layers, human experts have performed manual error detection and correction. Toward the realization of high-accuracy automatic error detection method, this paper first applies Self-Organizing Map to the observation data for comprehensively understanding of the error characteristics, and then proposes a method for error detection based on Conditional Random Field. Experimental results showed that the proposed classification method based on CRF successfully detected observation errors with significantly better accuracy than the existing automatic quality control method.

DOI： 10.1007/978-3-319-13359-1_52

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：IGI Global  

Students engage in many learning activities outside of class but, it is not easy for them to learn on their own because they also need to identify what activities to perform, decide how long to engage in them, evaluate their progress, shift to other activities if needed and avoid distractions aside from others. This research designed and implemented a learning support tool called Sidekick, which used a retrospective approach to help students analyze and evaluate their own behavior so they can adjust it accordingly. The results showed that students benefitted from understanding their behavior more. It also showed how students' learning behavior changed over time and the differences in the type and amount of change between learning sessions according to students' level of autonomy. Less autonomous students seemed to improve less compared to highly autonomous students however, the system was able to encourage them to recall and self-evaluate which they might not have done without the system.

DOI： 10.4018/ijdet.2014100103

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）  

researchmap

担当区分：筆頭著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Japanese Society for Artificial Intelligence  

In this study, we extract earthquake co-occurrence patterns for investigating mechanical interactions in the affected areas. To extract seismic patterns, both co-occurrence among seismic events in the event sequence and distances between the hypocenters to find hot spots must be considered. Most previous researches, however, have considered only one of these aspects. In contrast, we utilized co-occurring cluster mining to extract seismic patterns by considering both co-occurrence in a sequence and distance between hypocenters. Then, we acquired affected areas and relationships between the co-occurrence patterns and focal mechanisms from the 2011–2012 hypocenter catalog. Some results were consistent with seismological literature. The results include highly affected areas that may indicate asperity, and change of focal mechanisms before and after the Tohoku Earthquake.

DOI： 10.1527/tjsai.29.493

Scopus

researchmap

記述言語：英語  

researchmap

担当区分：筆頭著者   記述言語：英語  

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：Institute of Electronics, Information and Communication, Engineers, IEICE  

In standard fighting videogames, users usually prefer playing against other users rather than against machines because opponents controlled by machines are in a rut and users can memorize their behaviors after repetitive plays. On the other hand, human players adapt to each other's behaviors, which makes fighting videogames interesting. Thus, in this paper, we propose an artificial agent for a fighting videogame that can adapt to its users, allowing users to enjoy the game even when playing alone. In particular, this work focuses on combination attacks, or combos, that give great damage to the opponent. The agent treats combos independently, i.e., it is composed of a subagent for predicting combos the user executes, that for choosing combos the agent executes, and that for controlling the whole agent. Human users evaluated the agent compared to static opponents, and the agent received minimal negative ratings. Copyright © 2014 The Institute of Electronics, Information and Communication Engineers.

DOI： 10.1587/transinf.E97.D.842

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）   出版者・発行元：WORLD SCIENTIFIC PUBL CO PTE LTD  

Students often face difficulty in self-directed learning scenarios (e.g., studying, research) because they need to control many aspects of the learning session. They need to decide what to learn, how long to perform a learning task, when to shift to a different learning task and manage distractions apart from others. We observed from our previous research that self-reflection and self-evaluation helped students manage their own learning. However, majority of the students only evaluated one or two major aspects of the learning session that they think needed to be changed or improved (e.g., need to spend less time in non-learning related activities, need to focus on only one learning task at a time). If students would look further into their learning session, they would discover more behaviors that also need to be re-evaluated. In this paper we discussed reinforcement learning-based methods for discovering good learning behavior which can be used by future systems to suggest to students possible ways to improve their behavior.

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）  

researchmap

記述言語：英語  

researchmap

記述言語：英語  

researchmap

記述言語：英語  

researchmap

記述言語：英語  

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）  

researchmap

担当区分：筆頭著者,　責任著者   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）  

This study proposes a distance metric learning method based on a clustering index with neighbor relation that simultaneously evaluates inter-and intra-clusters. Our proposed method optimizes a distance transform matrix based on the Mahalanobis distance by utilizing a self-adaptive differential evolution (jDE) algorithm. Our approach directly improves various clustering indices and in principle requires less auxiliary information compared to conventional metric learning methods. We experimentally validated the search efficiency of jDE and the generalization performance. © 2013 IEEE.

DOI： 10.1109/ICTAI.2013.66

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：Electrochemical Society Inc.  

We previously developed a technique by which to measure the mechanical damage of SOFC using the acoustic emission (AE) method. In the present paper, we applied an adapted Self-Organizing Map (SOM), which is an artificial neural network model, to produce a cluster map reflecting the similarity of AE events. The obtained map visualized the change in occurrence patterns of similar AE events, revealing six phases of damage progress. Moreover, we inferred mechanical interactions among components of SOFC from a series of AE events by our proposed data mining method called co-occurring cluster mining. Our methods provide a common foundation for a comprehensive damage evaluation system and a damage monitoring system. © The Electrochemical Society.

DOI： 10.1149/05701.0571ecst

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）  

Self-regulated learners have been shown to learn more effectively. However, it is not easy to become self-regulated because learners have to be capable of observing and evaluating their thoughts, actions and behaviors while learning. In this work, we used Q-learning to reveal the effectiveness or ineffectiveness of a learning behavior that carries over learning episodes. We also showed different types of effective learning behavior discovered and how they were differentiated. Providing learners with knowledge about learning behavior effectiveness can help them observe how strategy selection affects their performance and will help them select more appropriate strategies in succeeding learning episodes for better future performance. © 2013 Springer-Verlag Berlin Heidelberg.

DOI： 10.1007/978-3-642-39112-5-85

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）  

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）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1527/tjsai.27.121

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）   出版者・発行元：SPRINGER  

Time-interval sequential patterns provide information not only on frequently occurring items and the order in which they happen but also reveal the temporal dimension between successive items. Although time-interval data have been dealt with in the past - as single or multiple, regular or irregular, and/or with definite ranges, what we are proposing here is a data mining algorithm that allows multiple time intervals in a sequence that are irregular and more flexible by employing a clustering technique integrated in an Apriori-based algorithm. Clustering allows non-integral time values to be categorized effectively and efficiently and leads to the characterizations of time interval data. In light of our research on a smart space that aims to provide empathic support to its occupant, we aim to use our algorithm as tool when building various predictive models of human behavior. Behavior modeling is a persisting and compelling issue in the design of intelligent environments in order to anticipate user needs and provide timely system responses. Insensitive or untimely system responses solicit unfavorable user reception. As proof of concept, we used our algorithm to infer the behavior patterns of individuals in terms of their habitual paths and walk time, i.e., spots in the space that an individual would likely take coupled with walk duration intervals. Our smart space may then use these two parameters to create models of effective timely interactive support provisions.

Web of Science

researchmap

担当区分：責任著者   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）  

In this study, we research the mechanical correlations among components of solid oxide fuel cell (SOFC) by analyzing the co-occurrence of acoustic emission (AE) events which are caused by damage. Then we propose a novel method for mining patterns from the numerical data such as AE. The proposed method extracts patterns of two clusters considering co-occurrence between clusters and similarity within each cluster at the same time. In addition, we utilize the dendrogram obtained from hierarchical clustering for reduction of the search space. We applied the proposed method to AE data, and the damage patterns which represent the main mechanical correlations were extracted. We can acquire novel knowledge about damage mechanism of SOFC from the results. © 2012 Springer-Verlag.

DOI： 10.1007/978-3-642-30220-6_5

Scopus

researchmap

担当区分：筆頭著者,　責任著者   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）  

This paper proposes a methodology for introducing a neighborhood relation of clusters to the conventional cluster validity measures using external criteria, that is, class information. The extended measure evaluates the cluster validity together with connectivity of class distribution based on a neighborhood relation of clusters. A weighting function is introduced for smoothing the basic statistics to set-based measures and to pairwise-based measures. Our method can extend any cluster validity measure based on a set or pairwise of data points. In the experiment, we examined the neighbor component of the extended measure and revealed an appropriate neighborhood radius and some properties using synthetic and real-world data. © 2012 Springer-Verlag.

DOI： 10.1007/978-3-642-30217-6_30

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 Japan  

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）  

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）  

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：情報処理学会  

複雑系において，潜在的な相互作用により稀に急激な事象遷移を起こすことがある．本研究では，事象系列において比較的稀であるが，状態が遷移するときに共起する事象を抽出する Essential Event Extractor（E3） 法を提案する．E3 法では，自己組織化マップ （SOM） を符号化のためにベクトル量子化 （VQ） として，また共起グラフを生成するためにキーグラフを用いる．そして，VQ のトポロジマップから得られる発生密度推定を参照することで，共起グラフ上で事象遷移を抽出する．本論文では，燃料電池の損傷試験で観測される Acoustic Emission（AE） 事象系列に対して E3 法を適用し，燃料電池の損傷メカニズムの解明に向けた，有用な損傷事象系列を得た．Although sudden changes of the event phase in complex system may indicate potential essential forces, such events are rare. In the present paper, we propose an essential event extractor (E3) scheme to extract relatively rare but co-occurring event sequences in event transitions. In E3, the self-organizing map (SOM) is used as vector quantization (VQ) and KeyGraph as a co-occurrence graph. Afterwards, event transitions on the KeyGraph can be obtained by using occurrence density estimation on the topology map of VQ. We demonstrate the E3 using an acoustic emission (AE) event sequence observed during a damage test of fuel cells and obtain essential co-occurring damage sequences.

CiNii Books

researchmap

その他リンク： http://id.nii.ac.jp/1001/00073757/

担当区分：筆頭著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

We proposed the methodology of introducing topographic component to conventional clustering measures for the evaluation of the SOM using external criteria, i.e., class information. The topographic measure evaluates clustering accuracy together with topographic connectivity of class distribution on the topology space of the SOM. The topographic component is introduced by marginalization of basic statistics to the set-based measures, and by a likelihood function to the pairwise-based measures. Our method can extend any clustering measure based on set or pairwise of data points. The present paper examined the topographic component of the extended measure and revealed an appropriate neighborhood radius of the topographic measures. © 2011 Springer-Verlag Berlin Heidelberg.

DOI： 10.1007/978-3-642-21566-7_13

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）  

Although the design of our constructive adaptive user interface (CAUI) for an affect-based music compositional artificial intelligence has been modified on several fronts since the time it was introduced, what has become a persisting limitation of our research is the extent by which it should efficiently cover music theory effectively. This paper reports our initial investigation on the possible significant contribution of melodic motif in creating compositions that are more fluent and cohesive. From an initial collection of 10 melodic motifs from different musical pieces, we provided heuristic-based renditions to these melodic motifs, four for each one, and obtained a total of 50 melodic motifs. We asked 10 subjects to provide self-annotations of the affective flavor of these motifs. We then represented these motifs as first-order logic predicates and employed inductive logic programming for the CAUI to learn relations of user affect perceptions and music features. To obtain new compositions, we first used a genetic algorithm with a fitness function that is based on the induced relations for the CAUI to generate chordal tone variants. We then used probabilistic modifications for the CAUI to alter these chordal tones to become non-harmonic tones. The CAUI composed 60 new user-specific affect-based musical pieces for each subject. Our results indicate that the compositions differ significantly for only one pair of affect type when the subject evaluations of the CAUI compositions were compared using paired t-test. However, when we compared the subject evaluations of the quality of the melodies and of the musical pieces from when melodic motif variants were not considered, the improvement is significant with t-values of 5.86 and 6.33, respectively, for a significance level of 0.01. © 2011 IEEE.

DOI： 10.1109/KSE.2011.42

Scopus

researchmap

担当区分：筆頭著者,　責任著者   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 Japan Society of Mechanical Engineers  

The fuel cell is regarded as a highly efficient, low-pollution power generation system. In particular, Solid Oxide Fuel Cell (SOFC) has a high generation efficiency. However, a crucial issue in putting SOFC to practical use is the establishment of a technique for evaluating the deterioration. We previously developed a technique by which to measure the mechanical damage of SOFC using the Acoustic Emission (AE) method. In the present paper, we applied the kernel Self-Organizing Map (SOM), which is an extended neural network model, to produce a cluster map reflecting the similarity of AE events. The obtained map visualized the change in occurrence patterns of similar AE events, revealing four phases of damage progress. The methodology of the present study provides a common foundation for a comprehensive damage evaluation system and a damage monitoring system.

DOI： 10.1299/jee.6.499

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

Although sudden changes of the event phase in complex system may indicate underlying essential forces, such events are not frequent. In the present paper, we propose an essential event extractor (E3) scheme to extract relatively rare but cooccurring event sequences in event phase transitions. In E3, the self-organizingmap (SOM) is used as vector quantization (VQ) to encode non-symbolic events and Key-Graph as a co-occurrence graph. Afterwards, event transitions on the co-occurrence graph can be obtained by referring to an occurrence density estimation on the topology map of VQ. We demonstrate the E3 using an acoustic emission (AE) event sequence observed during a damage test of fuel cells and obtain reasonable and essential co-occurring damage sequences that exhibit mechanical effects. © Springer-Verlag Berlin Heidelberg 2011. © Springer-Verlag Berlin Heidelberg 2011.

DOI： 10.1007/978-3-642-19618-8_6

Scopus

researchmap

担当区分：筆頭著者,　責任著者   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）  

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  

Web of Science

researchmap

担当区分：筆頭著者,　責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

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）  

Data-centric affect modeling may render itself restrictive in practical applications for three reasons, namely, it falls short of feature optimization, infers discrete affect classes, and deals with relatively small to average sized datasets. Though it seems practical to use the feature combinations already associated to commonly investigated sensors, there may be other potentially optimal features that can lead to new relations. Secondly, although it seems more realistic to view affect as continuous, it requires using continuous labels that will increase the difficulty of modeling. Lastly, although a large scale dataset reflects a more precise range of values for any given feature, it severely hinders computational efficiency. We address these problems when inferring physiology-affect relations from datasets that contain 2-3 million feature vectors, each with 49 features and labelled with continuous affect values. We employ automatic feature selection to acquire near optimal feature subsets and a fast approximate kNN algorithm to solve the regression problem and cope with the challenge of a large scale dataset. Our results show that high estimation accuracy may be achieved even when the selected feature subset is only about 7% of the original features. May the results here motivate the HCI community to pursue affect modeling without being deterred by large datasets and further the discussions on acquiring optimal features for accurate continuous affect approximation. Copyright 2010 ACM.

DOI： 10.1145/1719970.1719974

Scopus

researchmap

記述言語：英語   出版者・発行元：Springer  

researchmap

担当区分：筆頭著者,　責任著者   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 Berlin Heidelberg  

DOI： 10.1007/978-3-642-01307-2_98

researchmap

担当区分：筆頭著者,　責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1527/tjsai.23.319

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：J.UCS Consortium  

DOI： 10.3217/jucs-014-07-1136

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：Macquarie Scientific Publishing  

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：SPRINGER-VERLAG BERLIN  

In recent years, it becomes vigorous to forecast a market by using machine learning methods. Since they assume that each trader's individual decisions do not affect market prices at all, most existing works use a past market data set. Meanwhile there is an attempt to analyze economic phenomena by constructing a virtual market simulator, where human and artificial traders really make trades. Since prices in the market are determined by every trader's decisions, it is more realistic and the assumption cannot be applied any more. In this work, we design and evaluate several reinforcement learners on a futures market simulator U-Mart (Unreal Market as an Artificial Research Testbed). After that, we compare our learner to the previous champions of U-Mart competitions.

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）  

Sensor-network technology is indispensable for constructing ubiquitous network infrastructures. Although information about adjacent relations between sensors is also very important for sensor networks, obtaining this information automatically without manual assistance is extremely difficult. Consequently, we propose a new methodology for constructing adjacent relations in sensor networks using an ant-colony optimization algorithm. This methodology can be used to automatically extract adjacent relations without using prepared sensor-location information or RFIDs to identify individual humans. We implemented a prototype system, and verified its basic effectiveness through simulations and an experiment using real data. © 2008 IEEE.

DOI： 10.1109/WIIAT.2008.143

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）  

Empathy is a learnable skill that requires experiential learning and practice of empathic ability for it to improve and mature. In the context of human-system interaction (HSI) this can mean that a system should be permitted to have an initial knowledge of empathy provision that is inaccurate or incomplete, but with this knowledge evolving and progressing over time through learning from experience. This problem has yet to be defined and dealt in HSI. This paper is an attempt to state an empathy learning problem for an ambient intelligent system to self-improve its empathic responses based on user affective states. ©2008 IEEE.

DOI： 10.1109/HSI.2008.4581435

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）  

We have proposed neural-network based visualization approach, called Sequence-based SOM (Self-Organizing Map) that visualizes transition of dynamic clusters by introducing the sequencing weight function onto the neuron topology. This approach mitigates the problems with a sliding window-based method. In this paper, we confirmed the properties of the proposed method via artificial data sets, and a real news articles data set by showing the topics' derivation and diversification/convergence. Visualization of cluster transition aids in the comprehension of such phenomena which come useful in various domains such as fault diagnosis and medical check-up, among others. © 2008 IEEE.

DOI： 10.1109/CIT.2008.4594648

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）  

researchmap

担当区分：筆頭著者,　責任著者   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/ICTAI.2007.94

Web of Science

researchmap

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

researchmap

担当区分：筆頭著者,　責任著者   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）  

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  

Web of Science

researchmap

担当区分：筆頭著者,　責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

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）  

It is beneficial to automatically acquire the sensor-network topology, which is geographical adjacency of sensors. We employed the Ant Colony Optimization (ACO) to solve this problem. We empirically validated the algorithm using the simulated and the real world sensor data.

DOI： 10.1109/INSS.2007.4297434

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）  

We consider a framework that learns human habitual behaviors from the data obtained by various kinds of sensors installed in an environment, so that the environment can interact with us based on those patterns. In this paper, we achieved extracting human behaviors by infrared sensor network as an initial step for the framework. Infrared sensor network is able to track us without putting an extra burden on us. Moreover it is able to collect long-term data. However, tracking with it has two problems, that is, link miss and incorrect link. In order to mitigate these problems, we propose the tracking method utilizing estimated "time distances" between sensors from movements' records. We have installed infrared sensor network in our laboratory, and validated the proposed tracking method by test courses. Afterwards, we confirmed that human behaviors can be extracted from longterm data.

DOI： 10.1109/INSS.2007.4297404

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）  

DOI： 10.1007/11893004_119

researchmap

記述言語：日本語  

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：Transducer Research Foundation  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.chb.2007.01.007

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：IOS Press  

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  

researchmap

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：機関テクニカルレポート，技術報告書，プレプリント等  

J-GLOBAL

researchmap

記述言語：日本語   掲載種別：機関テクニカルレポート，技術報告書，プレプリント等  

J-GLOBAL

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：書評論文，書評，文献紹介等  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

担当区分：筆頭著者   記述言語：日本語  

researchmap

担当区分：最終著者   記述言語：日本語   掲載種別：速報，短報，研究ノート等（学術雑誌）  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）  

researchmap

記述言語：日本語   掲載種別：速報，短報，研究ノート等（学術雑誌）  

researchmap

記述言語：日本語   掲載種別：会議報告等  

researchmap

担当区分：筆頭著者   記述言語：日本語   掲載種別：記事・総説・解説・論説等（学術雑誌）   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/1004/00001795/

記述言語：日本語  

researchmap

記述言語：日本語   掲載種別：速報，短報，研究ノート等（学術雑誌）  

researchmap

記述言語：日本語   掲載種別：会議報告等  

researchmap

記述言語：日本語   掲載種別：書評論文，書評，文献紹介等  

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（学術雑誌）  

CiNii Books

researchmap

担当区分：筆頭著者   記述言語：日本語   掲載種別：記事・総説・解説・論説等（学術雑誌）   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/1004/00008310/

記述言語：日本語   掲載種別：会議報告等  

CiNii Books

researchmap

記述言語：日本語   掲載種別：会議報告等   出版者・発行元：一般社団法人 人工知能学会  

DOI： 10.11517/jjsai.27.6_668

CiNii Books

researchmap

記述言語：日本語   掲載種別：記事・総説・解説・論説等（その他）  

researchmap

DOI： 10.5772/9174

researchmap

DOI： 10.1115/FuelCell2009-85238

researchmap

記述言語：日本語   会議種別：公開講演，セミナー，チュートリアル，講習，講義等  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：シンポジウム・ワークショップ パネル（指名）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

開催年月日： 2024年

researchmap

開催年月日： 2022年

researchmap

開催年月日： 2022年

researchmap

開催年月日： 2022年

researchmap

開催年月日： 2020年

researchmap

開催年月日： 2019年

researchmap

開催年月日： 2017年11月

記述言語：英語  

researchmap

開催年月日： 2017年

記述言語：日本語  

<p>大気中に浮遊している微小な液滴や固体粒子であるPM2.5は、人への健康被害だけでなく、酸性雨や地球温暖化といった環境問題にまで関係する。本発表では北京市での大気観測を例に、単一微粒子質量分析計によって測定した数百万個～数千万個のPM2.5粒子それぞれの質量スペクトルデータを用いた、PM2.5化学組成やその粒径分布、時間変動や発生源の解析を紹介し、PM2.5研究におけるビッグデータ分析の応用を示す。</p>

researchmap

開催年月日： 2016年11月

記述言語：英語  

researchmap

開催年月日： 2016年10月

記述言語：日本語  

researchmap

開催年月日： 2016年7月

記述言語：日本語  

researchmap

開催年月日： 2016年7月

記述言語：日本語  

researchmap

開催年月日： 2016年

記述言語：日本語  

researchmap

開催年月日： 2015年12月

記述言語：英語  

researchmap

開催年月日： 2015年2月

記述言語：日本語  

近年,コンピュータやセンサ技術の発達からデータをデータストリームとして連続的に獲得できる環境が増加している.しかし,現在これらの豊富なデータから十分に知識を獲得できているとは言えない.データストリームに対して学習手法を用いて知識の獲得を試みる際には,時間の経過と共に学習すべき概念が変化すること,つまりコンセプトドリフトに対応する必要がある.そのための方法の一つとしてコンセプトドリフト検出手法がある.これまで,コンセプトドリフト検出手法の研究は教師あり学習を中心に行われてきた.教師なし学習においても研究は行われ,一定の成果は収めたが計算量が大きいという欠点があった.そこで,本研究では,教師なし学習において計算量の少ないコンセプトドリフト検出手法を目指して,教師あり学習と信号処理において効果が確認されているコンセプトドリフト検出手法を応用し実験を行った.その結果,今後のコンセプトドリフト検出手法の開発において重要な知見を得た.

researchmap

開催年月日： 2015年

記述言語：日本語  

&lt;p&gt;本研究では、自身が開発した再生可能エネルギーの地域別最適化および評価ツールを用い、日本の全市区町村を対象にファジィ数理モデルによって再生可能エネルギーの組み合わせ最適解および評価指標の値を算出した。更に得られた解を用いてエネルギー需給や地域特性によるクラスタリングを行うことで、市区町村を類型化し、地域適合型のエネルギー計画策定に資する知見を得た。&lt;/p&gt;

researchmap

開催年月日： 2015年

記述言語：日本語  

researchmap

開催年月日： 2015年

記述言語：日本語  

researchmap

開催年月日： 2015年

記述言語：日本語  

researchmap

開催年月日： 2014年6月

記述言語：日本語  

本研究では,我々が以前に提案した共起クラスタマイニング(CCM)という手法の改良に取り組む.共起クラスタマイニングとは,複数のクラスタ間における共起性とクラスタ内の類似性を同時に考慮し,共起する2つのクラスタの範囲を決定する手法である.しかし,これまでクラスタ内の事象の時系列上の前後関係や生起間隔は簡単のため考慮していない.そこで今回は,新たにそれらを含めたクラスタ系列パターンの抽出を目指す.その中で,ベイズ推定を用いた手法を提案する.

researchmap

開催年月日： 2014年1月

記述言語：日本語  

データマイニングや機械学習において距離定義は,クラスタリングや判別学習の結果に多大な影響を与える.著者らは教師ありの大域距離学習方式として,大域的クラスタ妥当性指標に基づいた距離学習を行う方式を提案している.この方式の利点は,クラスラペルを教師データとして直接用いることができる点と,クラスタ近傍とクラスタ内部の評価を同時に行える点である.本研究では,上記の方式が実データに対してどの程度有効であるかを検討する.距離学習を行うことで良好なクラスタリング結果が得られることを確認する.

researchmap

開催年月日： 2014年

記述言語：日本語  

researchmap

開催年月日： 2013年3月

記述言語：日本語  

本研究では、波形や位置情報など特徴量で表される事象の系列から、系列データ上で互いに近接しており、かつ頻出する事象のペアである共起パターンを抽出する「共起クラスタマイニング」という手法を提案する。提案手法は、時系列上におけるクラスタ「間」の共起性と、特徴空間におけるクラスタ「内」の類似性を同時に考慮して、2つのクラスタの範囲を決定する手法である。まず、人工データを用いて単純な共起パターン抽出法(2段階法)と提案手法の評価を行った。次に、実データへの適用例として、燃料電池の損傷評価試験から得られた波形データと、2011年の東北地方太平洋沖地震の震源リストデータへ適用し、共起パターンの抽出を行った。

researchmap

開催年月日： 2012年9月

記述言語：日本語  

New energy conversion systems develop in environmental fields, and make its presence greater and greater. Data mining method was developed in order to investigate the mechanical and electrochemical degradations process of solid electrochemical devices under operating conditions. By using this method, it was possible to show and help the possibility that detailed degradation process can be evaluated visually and in real-time, even when applied to modules at the actual equipment level. On the other hand, as service science intends to scientifically reveal the mechanism and principle in order to create new energy devices, there emerges a lot of interdisciplinary approach of the brand-new field. While traditional marketing theories have provided frameworks of how to grasp, reach, and lure as many customers as possible, none of them can externalize the functional structure of any service knowledge, and offer methodology of handling them on computer. Aiming at computationally utilizing any service knowledge, we discuss service ontology by both analogies with functional ontology on ontology engineering, and by comparison with typical marketing theory. In addition, we introduce case study that are based on ontology, and show how to create knowledge by this theory.

researchmap

開催年月日： 2012年2月

記述言語：日本語  

本論文では，クラスタ間の近傍関係も考慮した大域的クラスタ妥当性指標に基づく距離学習法を提案する．提案法は，大域的クラスタ妥当性指標を適合度とした遺伝的アルゴリズム (GA) により，マハラノビス距離に基づく変換行列を学習する．評価実験では，妥当性指標の最適な平滑化半径の決定について示し，続いて距離学習により妥当性指標の値が改善することを確認すると共に，自己組織化マップ (SOM) による可視化を通じて提案法の有効性について考察する．This paper proposes a metric learning method based on the global cluster validity index that considers inter-cluster neighborhood relation. The proposed method learns a transformation matrix of Mahalanobis distance by genetic algorithm, where the global cluster validity index as the fitness function. The experiments reveals determination of the appropriate smoothing radius of the validity index, then confirms that the proposed distance metric learning improves the validity index. Also, we discuss the effectiveness of the proposed method via visualization by the self-organizing maps.

researchmap

開催年月日： 2012年1月

記述言語：英語  

researchmap

開催年月日： 2012年

記述言語：日本語  

researchmap

開催年月日： 2012年

記述言語：日本語  

researchmap

開催年月日： 2011年11月

記述言語：日本語  

燃料電池(特に固体酸化物燃料電池)は実用化に向けて研究が進んでいるが,物理的損傷の要因の特定が課題となっている.そこで,損傷時に生じる弾性波(AE)を計測し,分析することで要因の特定を行っている.本研究では,データマイニングの手法によりAEを分類し,さらに,AEの出現についてその共起関係を分析することで損傷パターンの抽出を行い,燃料電池における力学的相関関係を把握する研究を行った.本研究では,AEのような数値観測量のデータからパターンを抽出できるような新たな手法を提案する.提案手法は,2つのクラスタにおいてクラスタ間の共起性とクラスタ内の類似性を同時に考慮して,共起する2つのクラスタの範囲(要素)を決定する手法である.また,探索空間の削減のため,階層型クラスタリングによるデンドログラム上で探索を行っている.本手法をまずは人工データに適用して,パターン抽出について性能評価を行った.次に,燃料電池のAEデータに適用して損傷パターンの抽出を行った.得られた結果について,著者のうち2名の燃料電池の専門家によって評価を行い,燃料電池の物理的損傷について新たな知識を得ることが可能となった.

researchmap

開催年月日： 2011年1月

記述言語：日本語  

researchmap

開催年月日： 2011年

記述言語：日本語  

researchmap

開催年月日： 2011年

記述言語：日本語  

researchmap

開催年月日： 2010年3月

記述言語：英語  

Industry-standard opponents in fighting videogames do not adapt, making them uninteresting in the long run. To increase the level of entertainment that can be derived from playing fighting videogames against the computer, we propose an adapting agent. In this context "adapting" is interpreted as modifying the level of the opponent to match that of the user, and learning to fight against the fighting style of the user. The proposed agent utilizes mainly Profit-Sharing and Pattern-Mining for achieving adaptation. The proposed agent was developed and evaluated against static opponents with real users.Industry-standard opponents in fighting videogames do not adapt, making them uninteresting in the long run. To increase the level of entertainment that can be derived from playing fighting videogames against the computer, we propose an adapting agent. In this context "adapting" is interpreted as modifying the level of the opponent to match that of the user, and learning to fight against the fighting style of the user. The proposed agent utilizes mainly Profit-Sharing and Pattern-Mining for achieving adaptation. The proposed agent was developed and evaluated against static opponents with real users.

researchmap

その他リンク： http://id.nii.ac.jp/1001/00068064/

開催年月日： 2010年2月

記述言語：日本語  

近年，燃料電池に対する関心は非常に高まっている．中でも固体酸化物燃料電池 （SOFC） は高い発電効率を持ち実用化が期待される発電器であるが，その実用化に向けては運転中に発生する劣化や損傷の診断が必須となる．これに対し現在，損傷の際に放出される弾性波を波形信号として取得することで物理的な損傷については観測が可能となった．この波形信号について，我々はデータマイニングの手法を用いて自動的に知識を抽出し，SOFC の損傷メカニズムを特定する研究を行ってきた．本研究は，損傷の発生する時系列に着目し，頻繁に発生する波形信号の系列パターンを抽出することで，実際にどのような損傷が起こっているのか，どのような損傷に時間的相関性があるのかを推定した．パターンの抽出においては，従来の記号データに対する系列パターン抽出手法を，非記号データである AE に適用できるように拡張を行った．Fuel cell is one of the energy generation systems of which practical use is expected. Especially, Solid Oxide Fuel Cell (SOFC) has high generation efficiency. However it is an extremely important problem that deterioration occurs while SOFC is under operation. A technique measuring the mechanical damage of SOFC has been previously developed by using Acoustic Emission (AE) method. In this research, we aim to specify the damage patterns of SOFC from AE waves by utilizing Sequential Pattern Mining method. However, it is difficult to apply conventional Sequential Pattern Mining to AE data directly because each AE wave is "non-symbolic data". Therefore, we extended Sequencial Pattern Mining so that it can apply to non-symbolic data.

researchmap

その他リンク： http://id.nii.ac.jp/1001/00068009/

開催年月日： 2010年1月

記述言語：日本語  

本稿では，き裂，摩擦や衝突音など損傷に関わるAcoustic Emission（AE）信号に対して，カーネルSOMのマップ上での隣接性を考慮した分類性能を評価した．AE信号群の周波数スペクトルの分布間の距離として，確率分布間の距離に基づくKullback-Leibler（KL）カーネルを用いた．また，SOMのマップ上での隣接性を考慮して，クラスタ純度やF値による通常のクラスタリング尺度を拡張した．複数の模擬データを用いて，いくつかの標準的なカーネルを用いたSOMや通常型SOMと比較した結果，KLカーネルを用いた場合にF値の観点で最も良い性能を示すことを確認した．また可視化結果とあわせて，クラスの分離性と密集性ついて考察した．We evaluated clustering perfomance of Kernel SOM considering adjacency within the obtained map upon Acoustic Emission (AE) waves involved in damage such as crack, friction and collision. Here, we employed Kullback-Leibler (KL) kernel that is based on a distance between probability distributions as a distance between frequency spectrum distributions. Also standard clustering measures, e.g., cluster purity and F-measure, are extended so as to consider adjacency within the map obtained by SOM. Using simulated AE data sets, we confirmed the KL kernel performs the best among the several standard kernels in terms of F-measure. Also we discussed about separability and density of classes together with the visualized maps.

researchmap

その他リンク： http://id.nii.ac.jp/1001/00067622/

開催年月日： 2010年1月

記述言語：英語  

researchmap

開催年月日： 2010年1月

記述言語：英語  

researchmap

開催年月日： 2009年10月

記述言語：英語  

researchmap

開催年月日： 2009年6月

記述言語：日本語  

researchmap

開催年月日： 2009年

記述言語：日本語  

researchmap

開催年月日： 2009年

記述言語：日本語  

researchmap

開催年月日： 2009年

researchmap

開催年月日： 2008年11月

記述言語：日本語  

researchmap

開催年月日： 2008年2月

記述言語：英語  

researchmap

開催年月日： 2008年

記述言語：日本語  

researchmap

開催年月日： 2007年12月

記述言語：日本語  

researchmap

開催年月日： 2007年6月

記述言語：英語  

researchmap

開催年月日： 2007年3月

記述言語：日本語  

researchmap

開催年月日： 2007年2月

記述言語：英語  

researchmap

開催年月日： 2006年12月

記述言語：英語  

researchmap

開催年月日： 2006年10月

記述言語：日本語  

researchmap

開催年月日： 2006年9月

記述言語：英語  

researchmap

開催年月日： 2006年7月

記述言語：日本語  

本論文ではWebにおいて他のサイトとのリンクパターンが等しいサイトは内容的にも等しいのではないかと考え，リンクパターンが等しいサイト間で比較を行うことでWebページから重要語抽出する手法を提案する．まず，CONCORでサイト間のリンク関係を表すネットワークをクラスタに分割し，提案手法によってそれぞれのサイトに対する類似サイトを特定する．そして，CONCORにより分割された同一クラスタ内のサイト間に共通して出現する名詞とそれぞれのサイトとそれらに対する類似サイト間に共通して出現する名詞の重み付けを補正する．リンク構造を言語処理に反映することで重要語抽出をおこない，従来手法よりよい結果が得られた．In this paper, we propose a framework to extracting significant words in the Web using link structure.We use the global method in social network 'CONCOR' for link analysis. This is based on the assumption that if the link patterns of two sites and links are the same, then these two sites also contain the same in contents. In the first phase, the whole network consisting of sites are divided into clusters using CONCOR.Subsequently, by using the method we propose, we identify similarity sites.Comparing a site with other sites in the same cluster and with the similarity sites for the site, we assign higher weights to nouns that exist in two sites in common.By using link analysis to language processing, we could discover significant words.

researchmap

その他リンク： http://id.nii.ac.jp/1001/00050183/

開催年月日： 2006年5月

記述言語：日本語  

大規模文章群からのトピック自動摘出は文章群全体像の把握や文盲分類などに有用であるとして、研究が行われている 。しかしここでは機械摘出トピックの評価は重要な課題であるが、人手によってトピックが分類された文章群が得られたとしても、トピック表現の多様性によりトピック間の対応付けが困難となるため、単純に機械摘出トピックと比較評価できない。そこで、本稿では、潜在的意味解析などによるベクトル表現可能なトピック抽出法を対象として、人手によってトピック分類され文章た群を用いて抽出トピックの解釈可能度を定量的定に評価することを試みた・日本語および英語の新聞記事などから機械摘出したトピック群にて提案評価法の妥当牲を検証した。Automatic topic extraction from a large number of documents is useful to figure out an entire picture of the documents or to classify the documents. Here, it is an important issue to evaluate the automatically extracted topics, however, even if manually-1abeledd ocuments are obtained, it is impossible to compare automatically aud manually derived topics due to complexity and uncertainty of the topics' structure. A8 the objective is vector representable topic extractions such as Latent Semantic Analysis, in this paper we tried to evaluate the interpretability of automatically extracted topics using the inanulally-1abeled documents. WQ validated the proposed evaluation method using topics extracted from Japanese and English news articles.

researchmap

その他リンク： http://id.nii.ac.jp/1001/00033150/

開催年月日： 2005年8月

記述言語：日本語  

ユビキタス環境下で、人間と環境との新しいインタラクションの枠組みを構築するのに有用な、時系列データからのセンサー隣接関係の可視化手法を提案する。大規模センサーネットワーク環境に及ぼす人間の習慣的行動パターンなどを抽出/学習するとともに、各種インタラクションデバイスを設置して適切なタイミングかつ適切な場所にて対象の各人にインタラクションを実行できるようなユビキタス環境の実現には、入念な事前設定を不要とする。センサーネットワークの自動コンフィグレーションが必要になると思われる。本論では実験的に多くのセンサーノードを設置した環境で、センサーネットワークの論理的隣接環境を、地理的情報無しでも、時系列発火データのみから抽出し、可視化する試みについて述べる。We propose the visualization method of sensor's relationship from chronological order of sensor data, without elaborate prior configuration or geometrical information of sensors. It is useful to build a frame of new interaction in ubiquitous environment. We extract a human custom action pattern to give it to sensor network environment and learn it, and ubiquitous environment offering appropriate service to is demanded. It seems that automatic configuration of a massively sensor network becomes need for ubiquitous environment realization. In this paper, we describe a trial to extract relationship of the experimental sensor network and visualized it.

researchmap

その他リンク： http://id.nii.ac.jp/1001/00050228/

開催年月日： 2005年8月

記述言語：日本語  

researchmap

開催年月日： 2004年12月

記述言語：英語  

researchmap

開催年月日： 2002年11月

記述言語：英語  

researchmap

記述言語：英語   会議種別：ポスター発表  

researchmap

記述言語：英語   会議種別：ポスター発表  

researchmap

記述言語：英語   会議種別：ポスター発表  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：ポスター発表  

researchmap

記述言語：日本語   会議種別：公開講演，セミナー，チュートリアル，講習，講義等  

researchmap

記述言語：日本語   会議種別：公開講演，セミナー，チュートリアル，講習，講義等  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：公開講演，セミナー，チュートリアル，講習，講義等  

researchmap

記述言語：日本語   会議種別：シンポジウム・ワークショップ パネル（指名）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：ポスター発表  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（招待・特別）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：メディア報道等  

researchmap

記述言語：日本語   会議種別：公開講演，セミナー，チュートリアル，講習，講義等  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：公開講演，セミナー，チュートリアル，講習，講義等  

researchmap

記述言語：英語   会議種別：公開講演，セミナー，チュートリアル，講習，講義等  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：ポスター発表  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：ポスター発表  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：ポスター発表  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：ポスター発表  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（招待・特別）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：英語   会議種別：口頭発表（招待・特別）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（招待・特別）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap

記述言語：日本語   会議種別：口頭発表（一般）  

researchmap