Haruhiko Kimura
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Haruhiko Kimura has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Takashi Oyabu
NIHONKAI International Exchange Center
13 shared publications Shigeki Hirobayashi
Toyama University
5 shared publications- Munehiro Nakamura
Graduate School of Natural Science and Technology, Kanazawa University
3 shared publications Hiromi Ban
2 shared publications- Makoto Koshino
Department of Electronics and Information Engineering, National Institute of Technology, Ishikawa College
2 shared publications Satoshi Sakamori
2 shared publications
Latest Publications
- 17 October 2017
- Published by DEStech Publications in DEStech Transactions on Engineering and Technology Research
- No. amsm
Metrical Characteristics of English Interviews
An “interview” is more specific way of talking, and it is the technique to gain the particular data effectively which the interviewer wants to know through the conversation. In this paper, English interviews: Larry King Live on CNN were metrically analyzed. Larry King Live is one of the CNN’s highest-rated shows, and Mr. King is regarded as the first American talk show host to have a worldwide audience. For comparison, English news materials from CNN Live Today as well as the inaugural addresses of the three U.S. Presidents were analyzed. In short, frequency characteristics of character- and word-appearance were investigated with a program written in C++. In this analysis, an approximate equation of an exponential function [y = c * exp(-bx)] was used to educe the characteristics of each material. Moreover, to obtain the difficulty-level, the percentage of the American basic vocabulary as well as the value of K-characteristic was calculated. As a result, it was clearly shown that the interviews have the same tendency as English journalism in character-appearance. Moreover, it was shown quantitatively that the interviews are a little easier to listen than CNN news.
- 17 October 2017
- Published by DEStech Publications in DEStech Transactions on Engineering and Technology Research
- No. amsm
Metrical Characteristics of English Interviews
An “interview” is more specific way of talking, and it is the technique to gain the particular data effectively which the interviewer wants to know through the conversation. In this paper, English interviews: Larry King Live on CNN were metrically analyzed. Larry King Live is one of the CNN’s highest-rated shows, and Mr. King is regarded as the first American talk show host to have a worldwide audience. For comparison, English news materials from CNN Live Today as well as the inaugural addresses of the three U.S. Presidents were analyzed. In short, frequency characteristics of character- and word-appearance were investigated with a program written in C++. In this analysis, an approximate equation of an exponential function [y = c * exp(-bx)] was used to educe the characteristics of each material. Moreover, to obtain the difficulty-level, the percentage of the American basic vocabulary as well as the value of K-characteristic was calculated. As a result, it was clearly shown that the interviews have the same tendency as English journalism in character-appearance. Moreover, it was shown quantitatively that the interviews are a little easier to listen than CNN news.
- Open Access
- 9 January 2017
- Published by Wiley in Electronics and Communications in Japan
- Vol. 100 (2) , 12-24
A Mood Prediction System for Preventing Diseases Using Biological Information and Weather Information
- 9 January 2017
- Published by Wiley in Electronics and Communications in Japan
- Vol. 100 (2) , 12-24
Open AccessA Mood Prediction System for Preventing Diseases Using Biological Information and Weather Information
- 28 December 2015
- Published by David Publishing Company in J. of Tourism and Hospitality Management
- Vol. 3 (6)
Text Data Mining of English Materials for Tourism
- 28 December 2015
- Published by David Publishing Company in J. of Tourism and Hospitality Management
- Vol. 3 (6)
Text Data Mining of English Materials for Tourism
- Open Access
- 1 February 2015
- Published by Springer Nature in SpringerPlus
- Vol. 4 (1) , 39
An experimental result of estimating an application volume by machine learning techniques
- 1 February 2015
- Published by Springer Nature in SpringerPlus
- Vol. 4 (1) , 39
Open AccessAn experimental result of estimating an application volume by machine learning techniques
- Open Access
- 2 October 2013
- Published by Springer Nature in BioData Mining
- Vol. 6 (1) , 16
LVQ-SMOTE – Learning Vector Quantization based Synthetic Minority Over–sampling Technique for biomedical data
Over-sampling methods based on Synthetic Minority Over-sampling Technique (SMOTE) have been proposed for classification problems of imbalanced biomedical data. However, the existing over-sampling methods achieve slightly better or sometimes worse result than the simplest SMOTE. In order to improve the effectiveness of SMOTE, this paper presents a novel over-sampling method using codebooks obtained by the learning vector quantization. In general, even when an existing SMOTE applied to a biomedical dataset, its empty feature space is still so huge that most classification algorithms would not perform well on estimating borderlines between classes. To tackle this problem, our over-sampling method generates synthetic samples which occupy more feature space than the other SMOTE algorithms. Briefly saying, our over-sampling method enables to generate useful synthetic samples by referring to actual samples taken from real-world datasets.
- 2 October 2013
- Published by Springer Nature in BioData Mining
- Vol. 6 (1) , 16
Open AccessLVQ-SMOTE – Learning Vector Quantization based Synthetic Minority Over–sampling Technique for biomedical data
Over-sampling methods based on Synthetic Minority Over-sampling Technique (SMOTE) have been proposed for classification problems of imbalanced biomedical data. However, the existing over-sampling methods achieve slightly better or sometimes worse result than the simplest SMOTE. In order to improve the effectiveness of SMOTE, this paper presents a novel over-sampling method using codebooks obtained by the learning vector quantization. In general, even when an existing SMOTE applied to a biomedical dataset, its empty feature space is still so huge that most classification algorithms would not perform well on estimating borderlines between classes. To tackle this problem, our over-sampling method generates synthetic samples which occupy more feature space than the other SMOTE algorithms. Briefly saying, our over-sampling method enables to generate useful synthetic samples by referring to actual samples taken from real-world datasets.
- Open Access
- 1 January 2012
- Published by Japanese Society for Artificial Intelligence in Transactions of the Japanese Society for Artificial Intelligence
- Vol. 27 (3) , 213-222
An Estimation of Difficulty for Academic Books using Reviews
A collaborate filtering has been generally used as a method which recommends items to customers. However, recommending academic books, it need to consider difficulty of them and individual amount of knowledge as well as user's preference. If the recommendation method considers only user's preference, they might regret after buying or reading recommended book because it won't match user's appropriate level. In this paper, we focus on academic books and propose a method which estimates the difficulty of academic books using user's reviews. Estimating difficulty of books will support users to search and recommend academic books that match user's skill. Moreover, we evaluated applying our method to academic text books about C programming Language. We verified that our method is more effective than traditional methods for academic books
- 1 January 2012
- Published by Japanese Society for Artificial Intelligence in Transactions of the Japanese Society for Artificial Intelligence
- Vol. 27 (3) , 213-222
Open AccessAn Estimation of Difficulty for Academic Books using Reviews
A collaborate filtering has been generally used as a method which recommends items to customers. However, recommending academic books, it need to consider difficulty of them and individual amount of knowledge as well as user's preference. If the recommendation method considers only user's preference, they might regret after buying or reading recommended book because it won't match user's appropriate level. In this paper, we focus on academic books and propose a method which estimates the difficulty of academic books using user's reviews. Estimating difficulty of books will support users to search and recommend academic books that match user's skill. Moreover, we evaluated applying our method to academic text books about C programming Language. We verified that our method is more effective than traditional methods for academic books
- 1 January 2012
- Published by Institute of Electrical Engineers of Japan (IEE Japan) in IEEJ Transactions on Sensors and Micromachines
- Vol. 132 (6) , 154-158
The Relationship between Chlorophyll Fluorescence Parameter (Fv/Fm) and Frequency Component of Plant Bioelectric Potential in Spraying Chemical Herbicides
- 1 January 2012
- Published by Institute of Electrical Engineers of Japan (IEE Japan) in IEEJ Transactions on Sensors and Micromachines
- Vol. 132 (6) , 154-158
The Relationship between Chlorophyll Fluorescence Parameter (Fv/Fm) and Frequency Component of Plant Bioelectric Potential in Spraying Chemical Herbicides
- 7 April 2004
- Published by Wiley in Systems and Computers in Japan
- Vol. 35 (6) , 89-97
Improving heuristic function of cost-based abduction system using real-time heuristic search
- 7 April 2004
- Published by Wiley in Systems and Computers in Japan
- Vol. 35 (6) , 89-97
Improving heuristic function of cost-based abduction system using real-time heuristic search
- 1 January 2001
- Published by Institute of Electrical Engineers of Japan (IEE Japan) in IEEJ Transactions on Sensors and Micromachines
- Vol. 121 (6) , 343-350
Recognition of Residential Behavious Using Odor Sensor Characteristics
Recognition rate of human behaviors in a residence is improved using an association rule, which is widely used in the field of the data mining. In the present study, plural odor-sensor characteristics are adopted to introduce the features of the behaviors. The rules in the database with positive instances, which are the real data, can be derived in this method. On the other hand, the negative instances are also focused to introduce various kinds of behavior-features. These instances express error data in the recognition process. First, we propose a method by which the positive and negative rules can be derived from the both kinds of instances. And we supose that the recognition results can be obtained by the rules. Secondarily, we perform evaluation experiments and compare the recognition rates of the LBG clustering with those of the proposed method. As for the results, we achieved the following results. (1) The recognition rate can be improved by the proposed method except the case of lack of the negative instances and the case of category in which the same activities are included as the positive and negative instances. (2) By analyzing the activity history, which the resident recorded his behaviors, we confirmed that the positive and the negative rules could correct the recognition results. The proposed method is useful to recognize the human behaviors in the residence
- 1 January 2001
- Published by Institute of Electrical Engineers of Japan (IEE Japan) in IEEJ Transactions on Sensors and Micromachines
- Vol. 121 (6) , 343-350
Recognition of Residential Behavious Using Odor Sensor Characteristics
Recognition rate of human behaviors in a residence is improved using an association rule, which is widely used in the field of the data mining. In the present study, plural odor-sensor characteristics are adopted to introduce the features of the behaviors. The rules in the database with positive instances, which are the real data, can be derived in this method. On the other hand, the negative instances are also focused to introduce various kinds of behavior-features. These instances express error data in the recognition process. First, we propose a method by which the positive and negative rules can be derived from the both kinds of instances. And we supose that the recognition results can be obtained by the rules. Secondarily, we perform evaluation experiments and compare the recognition rates of the LBG clustering with those of the proposed method. As for the results, we achieved the following results. (1) The recognition rate can be improved by the proposed method except the case of lack of the negative instances and the case of category in which the same activities are included as the positive and negative instances. (2) By analyzing the activity history, which the resident recorded his behaviors, we confirmed that the positive and the negative rules could correct the recognition results. The proposed method is useful to recognize the human behaviors in the residence