論文 - 山森 一人
-
An Automatic Model Building for Screening Functional Foods with GP 査読あり
Kunihito Yamamori, Ikuo Yoshihara, Mayumi Kamiguchi, Kazuo Nishiyama
ICROS-SICE International Joint Conference 2009 3679 - 3684 2009年8月
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)
-
GPを用いたバイオマーカ発現量からの機能性食品の生理活性値推定
上口真由美,山森一人,吉原郁夫,永濱清子
情報処理学会研究報告(バイオ情報学) 2009 ( 25 ) 21 - 24 2009年3月
記述言語:日本語 掲載種別:研究論文(その他学術会議資料等)
-
Feature extraction of protein expression levels based on classification of functional foods with SOM 査読あり
Tamon Fukushima, Kunihito Yamamori, Ikuo Yoshihara, Kiyoko Nagahama
Artificial Life and Robotics 13 ( 2 ) 543 - 546 2009年3月
記述言語:英語 掲載種別:研究論文(学術雑誌)
-
GPを用いたバイオマーカ発現量からの機能性食品の生理活性値推定
上口 真由美, 山森 一人, 吉原 郁夫, 永濱 清子
研究報告バイオ情報学(BIO) 2009 ( 25 ) 21 - 24 2009年2月
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:一般社団法人情報処理学会
GPを用いたバイオマーカ発現量からの生理活性値推定法について報告する。HepG2細胞に機能性既知の精製された食品成分を作用させてバイオマーカ発現量を測定し、GPによりモデル式を自動生成した。食品抽出物を同細胞に作用させたときのバイオマーカ発現量をこのモデル式に入力し、抗酸化ストレス・抗血管新生活性等の生理活性値の推定を行った。This research reports a new procedure to estimate physiological activity values of foods from protein expression levels using Genetic Programming(GP). Protein expression levels and physiological activities by refined constituents of foods are mesured in advance and they are used to make a model by GP. Then physiological activity values of food extracts are estimated by the above model,from their protein expression levels only.
-
Development of Physiological Activity Estimation Method of Foods Using Amplitude Extended Neural Networks 査読あり
T.Kuno, M.Kamiguchi, K.Yamamori, I.Yoshihara, K.Nagahama
Proceedings of The Fourteenth International Symposium on ARTIFICIAL LIFE AND ROBOTICS 658 - 661 2009年2月
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)
-
Togo S., Yamamori K., Yoshihara I., Nagahama K.
Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08 853 - 856 2008年12月
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:Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08
This paper addresses a novel method to estimate physiological activities of functional foods from protein expression levels using Bayesian classifier. The procedure of classification is as follows. Firstly, making clear relations between physiological activity values of functional foods and protein expression levels using simple regression analysis, secondly constructing Bayesian classifier to classify constituents of foods into three classes; "positive", "neutral" or "negative". Finally, giving the protein expression levels of constituents of foods to be examined to the Bayesian classifier and finding the class which the foods will belong to. Physiological activity of the foods is expressed by the above-mentioned class. We tackle to estimate five physiological activities by using expression levels of thirteen proteins. ©ISAROB 2008.
-
Feature extraction of protein expression levels based on classification of functional foods with SOM
Fukushima T., Yamamori K., Yoshihara I., Nagahama K.
Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08 849 - 852 2008年12月
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:Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08
We investigate relations between physiological activities and the protein expression levels of functional foods using Self-Organizing Map (SOM). The input vectors to SOM are composed of the protein expression levels and the physiological activity. A competitive node has two kinds of weights; one is for protein expression levels, the other is for physiological activity. A winner node is decided only by the weights for protein expression levels, and all weights in each node are updated. Each node has artificially generated value of physiological activity. The nodes can he categorized by the above-mentioned physiological activity. The well-trained SOM gives us some suggestions about relations between physiological activities and the protein expression levels. ©ISAROB 2008.
-
Variable-length segmental transmission line and its design guidelines
Koizumi N., Hayashi K., Yasunaga M., Yamamori K., Yoshihara I.
Artificial Life and Robotics 12 ( 1-2 ) 214 - 218 2008年12月
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:Artificial Life and Robotics
Waveform distortion is a serious problem in higher-frequency signals on printed circuit boards (PCBs). To overcome this problem, we have already proposed the segmental transmission line (STL) method, which divides transmission lines into several segments with different line widths. Each line width is adjusted to make the reflection noises cancel each other out in order to minimize the signal distortion. In this research, we first analyze the waveform reshaping mechanism of STL. Next, we apply STL to a dual in-line memory module (DIMM) clock-line for high-speed computers. The results give helpful guidelines for STL designs, which shows the efficiency of STL. © 2008 International Symposium on Artificial Life and Robotics (ISAROB).
-
BPA-PR法による階層型ニューラルネットワークの故障補償能力の向上
山森 一人, 森元 聡明, 吉原 郁夫
宮崎大學工學部紀要 37 275 - 280 2008年8月
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:宮崎大学
Some researchers have been proposed to implement neural networks into Wafer Scale Integration(WSI) to achieve fast learning. When neural networks are implemented into a WSI, it has to have a mechanism to avoid hardware defects. To compensate hardware defects, the partial retraining (PR) scheme has proposed. The performance of PR scheme depends on the weights in the neural network because PR scheme only adjusts the weights belonging to a neuron affected by the defects. In this paper, we propose back propagation with annealing scheme (BPA scheme) to improve defect compensation ratio. We show that BPA scheme achieved higher capability of defect compensation than that of conventional BP algorithm.
-
ニューラルネットワークによる生理活性値推定支援システムの開発
吉原 郁夫, 川西 千尋, 山森 一人
宮崎大學工學部紀要 37 287 - 292 2008年8月
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:宮崎大学
Recently westernized daily life has gradually changed our living circumstances and eating habits. Today most of the people are deeply concerning to functionality of foods, especially to the 3rd functionality. To know that, we had developed a proto-type system to estimate physiology activity values from protein expression levels. However, the proto- system is a little tough to operate for the analysts inexperienced in computers. This paper attempts to extend the proto-system with graphical user interface which is expectedly to let the system easier to use.
-
山森 一人, 園田 光, 吉原 郁夫, 相川 勝
宮崎大學工學部紀要 37 281 - 286 2008年8月
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:宮崎大学
Recently, computer graphics are used in various fields like movies, commercial films, etc. To generate high quality graphics, it takes much coumputing time for rendering process. So parallel computers such as cluster system are usually uesd to generate high quality graphics. In this paper, we discuss on the advantages and the disadvantages of there parallel ray-tracing algorithms; object-base parallelization, space-bace parallelization and pixel-base parallelization. Then we implement the parallel ray-tracing algorithm using pixel-base parallelization into a cluster system that has 44 processors, and evaluate the performance of the parallel ray-tracing algorithm.
-
山森 一人, 上口 真由美, 吉原 郁夫
宮崎大學工學部紀要 37 269 - 273 2008年8月
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:宮崎大学
Recent years, many researchers investigate functional foods which are reinforced useful constituents for human. These researches need to measure the physiological activities when a constituent is given to cells. However, it takes much time because the measurement of physiological activities needs many manual operations. Therefore, development of a system which can estimate the physiological activities of functional foods in short time is required. This research proposes a procedure to estimate physiological activity values of functional foods from protein expression levels. For this purpose we employ multiple regression analysis to derive an appropriate estimation function, and it is evaluated by using thirteen protein expression levels and various physiological activities.
-
Variable-length segmental transmission line and its design guidelines 査読あり
Naoki Koizumi, Kazuya Hayashi, Morotoshi Yasunaga, Kunihito Yamamori, Ikuo Yoshihara
Artificial Life and Robotics 12 214 - 218 2008年5月
記述言語:英語 掲載種別:研究論文(学術雑誌)
-
D-1-2 摂動法を併用したTSPの準最適解生成法(D-1. コンピュテーション,一般セッション)
黒田 正文, 岩切 淳一, 山森 一人, 吉原 郁夫
電子情報通信学会総合大会講演論文集 2008 ( 1 ) 2008年3月
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:一般社団法人電子情報通信学会
-
Feature Extraction of Protein Expression Levels Based on Classification of Functional Foods with SOM 査読あり
Tamon FUKUSHIMA, Kunihito YAMAMORI, Ikuo YOSHIHARA, Kiyoko NAGAHAMA
Proc. The Thirteenth International Symposium on Artificial Life and Robotics 2008 849 - 852 2008年1月
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)
-
Estimating Physiological Activities of Functional Foods from Protein Expression Levels Using Bayesian Classifier 査読あり
Shunichi TOGO, Kunihito YAMAMORI, Ikuo YOSHIHARA, Kiyoko NAGAHAMA
Proc. The Thirteenth International Symposium on Artificial Life and Robotics 2008 853 - 856 2008年1月
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 New Three-Level Tree Data Structure for Representing TSP Tours in the Lin-Kernighan Heuristic
NGUYEN Hung Dinh, YOSHIHARA Ikuo, YAMAMORI Kunihito, YASUNAGA Moritoshi
IEICE transactions on fundamentals of electronics, communications and computer sciences 90 ( 10 ) 2187 - 2193 2007年10月
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:一般社団法人電子情報通信学会
Lin-Kernighan (LK) is the most powerful local search for the Traveling Salesman Problem (TSP). The choice of data structure for tour representation plays a vital role in LK's performance. Binary trees are asymptotically the best tour representation but they perform empirically best only for TSPs with one million or more cities due to a large overhead. Arrays and two-level trees are used for smaller TSPs. This paper proposes a new three-level tree data structure for tour representation. Although this structure is asymptotically not better than the binary tree structure, it performs empirically better than the conventional structures for TSPs having from a thousand to three million cities.
-
A New Three-Level Tree Data Structure for Representing TSP Tours in the Lin-Kernighan Heuristic 査読あり
H.D.Nguyen, I.Yoshihara, K.Yamamori, M.Yasunaga
IEICE Transactions on Fundamentals E90-A ( 10 ) 2187 - 2193 2007年10月
記述言語:英語 掲載種別:研究論文(学術雑誌)
-
分割統治による Support Vector Machine の学習の高速化
山森 一人, 岩崎 隆, 吉原 郁夫, 相川 勝
宮崎大學工學部紀要 36 339 - 344 2007年8月
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:宮崎大学
Item Response Theory (IRT) is a method to analyze test and test items with assumpitions on the mathematical relationship between abilities and item responses. IRT is used for developing and refining test items in many fields of education. In this work, we carried out an IRT analysis by using data obtained from two types of programming tests. We applied the two parameter logistic model with item parameters aj and bj. The item parameter aj represents the discrimination power of the jth test item, and bj estimates ability, and this is caused partially by the insufficient number of test items.
-
揺らぎを含むタンパク質発現量からのニューラルネットワークによる食品の生理活性値の推定
山森 一人, 佃 晋輔, 吉原 郁夫
宮崎大學工學部紀要 36 345 - 350 2007年8月
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:宮崎大学
Recent years, many people is attracted to keep their health through foods. So a simple and speedy method is requested to know what kinds of effects are given by the ingredient in foods for human. However, the effect of ingredient in foods is very complex and have large variance, conventional estimation methods do not be applied for this task. In this report we propose a new method that combined with neural network and synthesis of training data from a few measured values in cells. Simulation results shows that our method succeeds to estimate the physiology activity values for fourteen ingredients in foods with acceptable accuracy.