Proceedings of the Fuzzy System Symposium
26th Fuzzy System Symposium
Displaying 151-200 of 299 articles from this issue
  • Yumi Endo, Jun Ozawa, Ichiro Kobayashi
    Session ID: TC4-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In these days, the global warming has become a very big issue for the human kind. We focus on a fact that the amount of CO2 production at home is much more larger than that by a car. Considering this fact, the objective of our study is to let family members recognized how much they use electricity in their everyday lives and make them aware of home ecology. As a first step for this, we propose a method to verbalize the amount and the trends of the use of electricity so that they can recognize them easily.
    Download PDF (269K)
  • Tomohiro Takagi, Akimitsu Mori, Yousuke Kaneko
    Session ID: TC4-5
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We will describe meaning representation, analogical reasoning, language computing environment and its applications using conceptual fuzzy sets.
    Download PDF (807K)
  • Masashi Sakai, Yutaro Tomoto, Masayoshi Kanoh, Tsuyoshi Nakamura
    Session ID: TD1-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    A method in which multi-valued decision diagrams (MDDs) are used to acquire robot action rules is proposed. Kanoh et al. have proposed a method, which uses multi-terminal binary decision diagrams (MTBDDs), to acquire robot action rules. But the variables of MTBDDs can only take values of 0 or 1; multiple variables are needed to represent a single joint angle. This increases the number of variables and the MTBDDs that represent the action rules become complex. Here, a method that uses MDDs, in which a single variable can take on multiple values, is proposed and experimental results are shown comparing MTBDDs and MDDs through simulations to acquire robot action rules.
    Download PDF (502K)
  • Hiroyuki INOUE, Chie INOUE
    Session ID: TD1-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Various colors are used for things that we use in daily life. A color combination is important for design and has a great influence on image or human kansei. When color combination of signboards is hammered away, it is necessary to consider not only the color combination of signboard but also blending in with the landscape. However, it is difficult to evaluate blending in with the landscape. Also, generation of color combination considering it is difficult. Therefore, in this study, the interactive genetic algorithm is used to the color combination support system of signboards. We try to construct the signboard color combination system with human subjective evaluation.
    Download PDF (2747K)
  • Yusuke Mori, Hidekazu Suzuki
    Session ID: TD1-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In the field of human interaction, soft actuator is needed for interpersonal operation. The soft actuator is required various skills such as high power (torque) drive, tenderness, angle control with high accuracy, and so on. Thus, we propose D3 (Dual Damper Drive) soft actuator. This actuator consists of two liner servomotor and elastic module, which connected at output shaft with antagonistic mechanism. We verify the performance of proposed system to perform torque control experiment using D3 actuator.
    Download PDF (257K)
  • Tomomi Hashimoto, Tomoyuki Sakurai, Yuji Udagawa, Yasuyuki Takakura, T ...
    Session ID: TD1-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In this paper, 1)we produce the walking simulator by using virtual reality, 2)compare young subjects and senior subjects, and 3)examines the usage of it. The system consists of PC, balance Wii board of Wii Fit, the steering wheel, and the cage. The subjects can walk in the virtual town by stepping on the board. November 2009 to May 2010, 9 senior citizens (5 male and 4 females) and 9 young people (9 male) used this system.
    Download PDF (415K)
  • Akira Ito, Tomohiro Yoshikawa, Takeshi Furuhashi, Ryoji Ikeda, Takahir ...
    Session ID: TD2-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In the field of marketing, questionnaires are often carried out in order to design a marketing strategy by analyzing acquired data. Association rule mining is one of the most famous data mining techniques in analysis for questionnaire data. The purpose of this method is to extract important regularities in data called association rules based on evaluation indicators concerning them. They are found to be very useful in practical application. However, a number of association rules are often discovered, which makes it difficult to grasp relations between rules and find interesting rules for users. This paper proposes a visualization method of relations between association rules with hierarchical graph structure. It applies the proposed method to an actual questionnaire data and shows that it supports users to grasp hierarchical relations between rules and search interesting rules for them.
    Download PDF (891K)
  • Shunsuke Akiguchi
    Session ID: TD2-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    It has been known that computers are not good at handling qualitative information, such as human impressions and sensitivities, while being good at quantitative mathematical calculations and handling large volume of information. However, in recent years, soft computing methods and sensitivity engineering are systematized. Then it is becoming possible to simulatively handle ambiguous information, such as human impressions and sensitivities with computers. Using these methods, in recent years, many researches have been conducted to relate media, such as images, and sensitivity information. A technique of classifying similar images using sensitivity information, such as impression words, and a technique of selecting sensitive characteristics from an image have been proposed. We, too, have created a system that automatically generates an image of pattern based on impression information input by the user. In this research, we would like to improve the former automatic pattern image generation system that we created, and aim to design a system, which can generate a pattern image that reflects the user's sensitivity better with a few elements.
    Download PDF (871K)
  • YUTARO TOMOTO, TSUYOSHI NAKAMURA, MASAYOSHI KANOH, TAKANORI KOMATSU
    Session ID: TD2-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Onomatopoeia refers to words that represent the sound, appearance, or voice of things, which makes it possible to create expressions that bring a scene to life in a subtle fashion. In this study, we propose an onomatopoeia thesaurus map to enable the construction of a map that visually confirms ``similarity relationships between a number of onomatopoetic words'' and ``similarity relationships between unknown onomatopoetic words and existing onomatopoetic words'', which are difficult to grasp from a conventional thesaurus. It is also possible to label objects with onomatopoetic words and visualize the similarity relationships between the objects on a map. In this paper, we introduce an example of labeling onomatopoetic words relating to the textures of sweets (desserts).
    Download PDF (1002K)
  • Kyoko Tsugioka, Hiromu Takahashi, Tomohiro Yoshikawa, Takeshi Furuhash ...
    Session ID: TD2-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Brain-Computer Interfaces (BCIs) are promising technologies to restore control and communication for severely paralyzed people such as those with amyotrophic lateral sclerosis, and they can be appealied to healthy people as well. The P300 speller is one of the BCI applications, which allows users to select letters just by thoughts. However, due to the low signal-to-noise ratio of the P300, signal averaging is often performed, which improves the spelling accuracy but degrades the spelling speed. Predictive text, implemented in the most mobile phones, enables users to spell with fewer key presses; thus, it could improve the spelling speed of the P300 speller. This paper implements a predictive text to the P300 speller, and examines the usability of the interface and how the spelling speed is improved.
    Download PDF (199K)
  • Toshio Shimokawa, Mitsuhiro Tsuji, Masanori Tanaka
    Session ID: TD3-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    The importance of team medical treatment in which doctors and co-medical colleagues (i.e., nurse, pharmacist, CRC, and so on) treat cancer together has been increasing. The nurse's main roles are to help with the cancer therapy and provide mental support to patients and families. We surveyed the concerned nurse's involvement in the cancer therapy using a questionnaire. We asked them the need for palliative care, importance of team medical treatment, etc. The purpose of our analysis was to explore factors that influence the nurse while providing palliative care. Multiple linear regression analysis is one of the most popular methods for performing data analysis of questionnaires. However, it cannot construct a model with an interaction effect and nonlinear structure. We adopted the rule ensemble (RE) method (Friedman and Popescu, 2008) because it can model nonlinear and interaction structures. Furthermore, in the RE method, the estimated model can be interpreted based on rules (base learner). Moreover, we developed a new graphical diagnosis for supporting the result. Therefore, we provided some influence factors about the intention of the nurse during palliative care and illustrated them using a graphical representation.
    Download PDF (494K)
  • Kenji Nakamura, Shigenori Tanaka, Kazufumi Otani, Yuhei Yamamoto
    Session ID: TD3-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In recent years, CGM such as bulletin board systems on which anyone can easily post information has been gaining attention due to the increased popularity of the Internet. Meanwhile, the increased circulation of illegal and harmful information has become a major issue. Particularly, posting advance notice of crimes to bulletin board systems has become a problem. As the messages to give advance notice of crimes include information on time and place, it is considered effective to provide information using GIS; however, the crime prevention maps published by local governments do not support handling of advance notice of crimes. To address this, we are conducting research aimed at building crime prevention maps by identifying advance notice messages of crimes on the Web, and then extracting and presenting information on them, using text mining. This paper proposes a method for identifying advance notice of crimes on bulletin board systems. We also demonstrate the effectiveness of the proposed method by comparing it with existing methods.
    Download PDF (537K)
  • Toshio Teraguchi, Shigenori Tanaka, Mitsuhiro Tsuji, Yuki Fukushima
    Session ID: TD3-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, the incidence of crime has decreased, but the interest in the safety of residents tends to be still high. And so, actions such as offering and gathering the crime information through Internet by the police and the local government have been performed. Therefore, we have studied about a method for recommendation of the security information considering circumstances. And, this paper focuses on methods for analysis of the characteristic information of the road using GIS. Many existing researches focused on functional character of road to analyze crimes. But, in these cases, there are few studies analyzed based on the relationships between road and the distribution of premises. However, by analyzing based on these characteristics of each road, it is suggested that can apply the obtained result to other similar roads. Therefore, this paper proposes a method for evaluating the local security by distinctive relationships between each road of crime scene and the distribution of the premises, and demonstrates the effectiveness of the proposed method.
    Download PDF (604K)
  • Masanori Ikebe, Shigenori Tanaka, Mitsuhiro Tuji
    Session ID: TD3-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, local governments' administrative services via the Internet has increased. Evaluation information on the administrative service came to be published in the Internet. However, the questionary investigation is a main current as the method of evaluating the administrative service, and it is difficult to acquire the silent majority published in the Internet. Therefore, the present study acquires various evaluations by collecting and accumulating evaluation information from the Internet. The satisfaction level of local residents is analyzed by comparison of collected evaluation information and administrative purposes.
    Download PDF (374K)
  • Yudai Tanaka, Michiyuki Hirokane, Yusuke Miyawaki, Wataru Shiraki
    Session ID: TD3-5
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, some huge natural disasters such as earthquake, typhoon and so on, have occurred at many area in the world, many people have often suffered terrible damages. Almost all the victims who do not understanding the location of evacuation center evacuate by depending on the hazard map, the evacuation sign and the local residents. But, it is actually difficult to make sure the hazard map after occurring the disaster. In the case of evacuation sign, in some places it showed course to evacuation center. But, in most parts of places it has been not shown route as it is now. And, another thing it is doubtful victims coolly make sure the evacuation sign in acute emergency as disaster. In this research, the influence of leaders during evacuation were considered by using the multi-agent-model based food evacuation system for decreasing the number of victims who failed to get out in time by the indication of the leader during evacuation to the escaping residence. Through some implementations by the multi-agent-model based food evacuation simulation, we discussed the necessary number of leader during evacuation and the actual arrangement of these leaders in the local area of Takatsuki city.
    Download PDF (1086K)
  • Daisuke Yamaguchi, Hirom Shioya, Ayahiko Niimi, Muneo Takahashi
    Session ID: TD4-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In this paper, we propose the Personal Software Process (PSP) Practice Support System efficiently to transmit only necessary programming capability and improvement for detailed design of Defect Types. This system can transmit programming to specific human among many software processes using a Agent technology. The system is also synthesized to do parallel and cooperative proposing internally. Applying the proposed method to a personal process-removing task, a flexible programming for quality of software. Agent studies the best transmission programming according to the makes planning in the personal process.
    Download PDF (655K)
  • Daisuke Yamaguchi, Fumiyo Katayama, Muneo Takahashi
    Session ID: TD4-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In this paper, we propose learning function for an agent acquires characteristic of this data aggregate autonomously so that an assortment technique by the artificial neural network which used revolving underwriting facility aggregation theory for defect and a reliability enhancement of a diagnostic information realizes high prediction accuracy when we use clinical practice diagnostic data from conventional research result by this case study, and it to be possible for diagnosis support system. We aim at three building of function generating the data set which necessary an agent used to the artificial neural network, and to learn medical diagnostic data (prostate cancer, kidney cancer, bladder cancer) by revolving underwriting facility aggregation theory handled at the urology department to be concrete.
    Download PDF (674K)
  • Tatsuya Minegishi, Ayahiko Niimi
    Session ID: TD4-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, because of increasing amount of data in the society, data stream mining targeting large scale data has attracted attention. The data mining is a technology of discovery new knowledge and patterns from the massive amounts of data, and what the data correspond to data stream is data stream mining. In this paper, we regard credit card transaction data as data stream and detect fraud use to apply decision tree algorithm to data stream mining. In applying decision tree algorithm, we propose new evaluative criterion for decision tree. We apply VFDT (Very Fast Decision Tree learner) algorithm to online decision tree construction. At first, we construct VFDT using those data and create detection of fraud use model. At second, we verify the classification accuracy of the model.
    Download PDF (168K)
  • Masahiro Inuiguchi, Ryuta Enomoto
    Session ID: TE1-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In order to analyze the distribution of individual mind-sets in a group, a hierarchical clustering of decision tables has been proposed. In the method, we can know how the individuals form clusters but the clusters are not always optimal in some criterion. In this paper, we develop non-hierarchical clustering techiniques for clustering of decision tables. We use a vector of rough membership values to represent individual opinon to a profile. Using rough membetship values, we develop fuzzy c-means methods for clustering decision tables. We examined the proposed methods in the clustering real world decision tables.
    Download PDF (129K)
  • Michinori Nakata, Hiroshi Sakai
    Session ID: TE1-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    A method of possible equivalence classes was developed to deal with missing values. To deal with imprecision of rough approximations that comes from missing values, the concepts of certainty and possibility were used. When an information table contains the missing values, two rough approximations, certain and possible ones, are obtained. The actual rough approximation is located between the certain and possible rough approximations. The method gives the same results as a method of possible worlds. This is a justification of the method of possible equivalence classes.
    Download PDF (232K)
  • Tetsuya Murai, Seiki Ubukata, Yasuo Kudo
    Session ID: TE1-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We extend Miyamoto's consideration on mappings based on multisets to to the cases from a point of view of hiearachical structure of sequences, multisets, and naive subsets.
    Download PDF (477K)
  • Yoshifumi Kusunoki, Masahiro Inuiguchi, Tetsuzo Tanino
    Session ID: TE1-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In order to analyse a decision table with a monotonicity between condition attributes and a decision attribute, the Dominance-based Rough Set Approach (DRSA) has been proposed. Moreover, to tolerate some degree of inconsistency, DRSA has been extended to Variable-Consistency DRSA (VC-DRSA) or Variable-Precision DRSA (VP-DRSA). Attribute reduction is one of the major tools of rough set approaches. In VC-DRSA and VP-DRSA, union-based reducts have been proposed. However, class-based reducts proposed in DRSA has not been studied in VC-DRSA and VP-DRSA. In this paper, we propose class-based reducts in VC-DRSA and VP-DRSA, and investigate relations among reducts.
    Download PDF (297K)
  • Shoji Hirano, Shusaku Tsumoto
    Session ID: TE2-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In this paper, we present a method for representing the granularity of data that have non-metric relational properties. Asymmetric, relational proximity is transcribed into a set of binary classifications, and the indiscernibility of objects are characterized by Jaccard coefficients. Objects are merged into single granule when we disregard fine discrimination knowledge by changing indiscernibility level. According to this, we built a dendrogram that represented hierarchy of granules. Using the brand swith data, we demonstrate that the hierarchy of granules produced by the discernibility-based approach could have reasonable correlation with the original dissimilarity matrix.
    Download PDF (214K)
  • Yasuo Kudo, Taichi Maruyama, Tetsuya Murai
    Session ID: TE2-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We propose a heuristic attribute reduction algorithm based on random selections of attributes. Rough set theory proposed by Pawlak provides theoretical foundations of set-theoretical approximation of concepts and logical reasoning from data, and attribute reduction is one of the most important research topics in the aspect of reasoning from data. The computational complexity of computing all reducts from the given data is NP-hard and there have been many heuristic attribute reduction algorithms; however, almost proposals compute just one candidate of reducts from the given data. In this paper, we propose a heuristic attribute reduction algorithm to compute reducts as many as possible, which is based on construction of small-sized decision tables from the given data by using random selection of attributes.
    Download PDF (151K)
  • Seiki Ubukata, Yasuo Kudo, Tetsuya Murai
    Session ID: TE2-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Agent control methods are generally required flexibility and robustness. As one of agent control methods to realize the flexibility, we proposed a variable neighborhood model in which an agent selects actions based on variable neighborhoods. In this paper, by its extension, we formulate a framework of multi-agent system based on variable neighborhood model. We also discuss interactions among agents in the model.
    Download PDF (136K)
  • Hitomi Okuma, Michinori Nakata, Hiroshi Sakai
    Session ID: TE2-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    A rule in a Deterministic Information System (DIS) is often defined by an implication T such that both support(T) > alpha and accuracy(T) > beta hold for the threshold values alpha and beta. In a Non-deterministic Information System (NIS), there are derived DISs due to the information incompleteness. A rule in a DIS was extended to either a rule in the lower system or a rule in the upper system in a NIS. This paper newly introduces a criterion, i.e., stability factor, into rules in a NIS. Rules in the upper system are classified according to the stability factor.
    Download PDF (240K)
  • Tetsuya Murai, Seiki Ubukata, Yasuo Kudo
    Session ID: TE3-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We consider morphology and image representation in Kripke models for non-classical logics.
    Download PDF (435K)
  • Tetsuya Murai, Seiki Ubukata, Yasuo Kudo
    Session ID: TE3-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We give a note on chord progression as sequences in Kripke models for non-classical logics. In this paper, we make some basic considerations on (1) chords as possible worlds and (2) chord progressions as transitions among chords as possible worlds.
    Download PDF (543K)
  • Katsuo Inoue, Akira Sekiguchi
    Session ID: TE3-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    These days, the visual impressions of user friendliness are paid to attention in the marketing field and the product design. From this background, though we focused the display and the operation panel in a product, we focused its visual impressions of user friendliness of the styling design only whose case study was a hair dryer in this paper. As an analysis method, we had the first attempt to extract the striking feature by the method to narrow samples using component scores of Principal component analysis. We have found that the original function of the target product such as "The hair seems to dry soon" ranked high and then the feature of each principal component was extracted by Rough sets. This result suggested a new aspect in the product concept making.
    Download PDF (340K)
  • Takeshi Shimmura, Takeshi Takenaka
    Session ID: TE4-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In restaurant businesses, design and operation mainly depend on experiences of skilled workers. Service engineering is a new research field to enhance productivity by scientific approach. This paper reports some studies conducted in a restaurant business; 1)designing cooking processes to enhance customer's value and productivity based on behavior observation of cooking, 2)improvement of restaurant processes by information sharing among staffs, and 3)analysis of employee's service skills at a point of contact with customers.
    Download PDF (2304K)
  • kazuo oshima
    Session ID: TE4-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Introduction of strategy outline Meiho applied in order to shift from contracting to fee business to meet transparency needs in business.
    Download PDF (426K)
  • Kenji Tomoda
    Session ID: TE4-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Drawing the speed of cars diagrammatically, gathering the dissociations between that and the ideal safe diagrammatic drawing which corresponded to the traffic situation, we indicated them. Being based on that indicator, we have instructed how to drive gently and smoothly. As a result, we were able to reduce both fuel for driving (CO2) and self-inflicted traffic accidents.
    Download PDF (2060K)
  • Yoshinobu Yamamoto, Yoshiyuki Nakamura, Muneo Kitajima
    Session ID: TE4-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    For the productivity improvement in service industries, it is important to repeat OADI Loop (Observation, Analysis, Design, and Implementation). However, in sightseeing areas, it is difficult to conduct continuous surveys such as pedestrian survey, tourists behaviors, etc. In this paper, we propose Service Survey by Services Method. Also we report a case study that this method has been applied in a hot spring resort in Japan.
    Download PDF (543K)
  • Tomoko Kashima, Shimpei Matsumoto, Hiroaki Ishii
    Session ID: TE4-5
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In this paper we develop a decision support system using fuzzy reasoning, where a menu is selected by the system. If we assume the menu planning for a hospital food service, home-delivery service, etc., an appropriate selection of the menu corresponding to an individual's preferences is required. In this case, fuzzy reasoning is used to determine the quantity of a nutrient, quantity of a certain dish, and which nutrient is important. Our proposed system suggests a menu choice that corresponds best to an individual for achieving a healthy and balanced nutrition.
    Download PDF (515K)
  • Mizuo Izawa, Yuichiro Taira, Motohiro Yokota, Makoto Nakamura
    Session ID: TF1-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We have predicted standing stock for phytoplankton using neural networks. Here, we discuss the dependence on frequency of observation and how frequently we need to observe water properties.
    Download PDF (257K)
  • Makoto Nakamura, Hiromitsu Ohta, Eiji Morimoto, Satoru Ezoe, Kazuki Ha ...
    Session ID: TF1-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    This study aims at constructing quality evaluation system of fresh fish. The appearance evaluation of dressed Tiger puffer by a cook of fish market was analyzed by statistical technique, and the relationship between the quality of the fish meat was examined. Some useful variables which show the color of fish body were extracted, and the fuzzy inference model which evaluates the quality of the fish using these variables was constructed. A moderate agreement between the simulated results and experiments is obtained, which implies the usefulness of the present system.
    Download PDF (386K)
  • Makoto Nakamura, Keisuke Matsumoto, Eiji Morimoto, Satoru Ezoe, Toshim ...
    Session ID: TF1-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    A model was developed to evaluate the appearance of fresh fish, the results of which were compared to the evaluations of professional fish auctioneers. Two different kinds of low chroma fishes (Horse-mackerel, Japanese Spanish mackerel), each differing with respect to the hue of fish body surface, were graded into five classes by experienced auctioneers. Analysis of appearance evaluations by auctioneers, combined with information on fish coloration and meat freshness, were used to develop three common color indexes for characterizing fish body surface. Fuzzy inference models were then constructed and evaluated using the indexes as explanatory variables, and the results obtained corresponded with the assessments of auctioneers with a certainty of more than 90%.
    Download PDF (373K)
  • Kenta Tanaka, Takashi Miyakuchi, Michiyuki Hirokane, Isao Hayashi
    Session ID: TF2-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In recent years, it demonstrate that concrete structure occur (happen) to decline quality by enlarged load, environment - temperature, humidity, carbon dioxide, acid rain and external pressure - materials, quality of work and it is recognized that maintenance and care are important. During maintenance and care, it is important to estimate and judge health of concrete structure. One of estimation and judgement of health, visual inspection of an expert is existent. Connecting visual information on the surface with state (situation) of declining quality inside of concrete is knowledge and experience of expert. This research focused TAM (Topographic Attentive Mapping) network as a method deal with visual information of expert. TAM network is neural network which models visual system processing from the retina down to first visual cortex and get same result which is human visual. We extract several amount of feature which based on cracking check pattern of surface of concrete board, and by using TAM network, have to master these expert knowledge by learning expert estimation and judgement. Moreover, by adopting Pruning Methods, we try to rise correct rate of TAM network and we extract special features which are factors of expert estimation and judgement from knowledge of learn.
    Download PDF (762K)
  • Naoki Motomura, Yasutoshi Nomura, Hitoshi Furuta
    Session ID: TF2-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, structural damage assessment methods using chaotic excitation have been proposed to detect damage location of civil structures as an important baseline method that can detect minor damages. Baseline methods need to prepare such information as response data of objective structure under intact condition. It is difficult to apply baseline methods to the structure without information of the baseline. To overcome the obstacle, this study attempts to develop a baseline-free damage detection method using chaotic excitation. The proposed method evaluates the spread of local areas of attractor as a feature. Through numerical simulations it is proven that the proposed method can make it possible to detect the damage location without using baseline data.
    Download PDF (294K)
  • Isao Hayashi, Yinlai Jiang, Masanao Hara, Shuoyu Wang
    Session ID: TF2-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    There is movement instruction as nonverbal communication in the human communication. However, we have infinite pattern for movement instructions by our personality. On the other hand, the singular spectrum analysis method has been proposed as analytical method for time-series data. In this paper, we propose a method for acquiring knowledge from time-series gesture data using singular spectrum analysis. We acquire knowledge in terms of gesture characteristic by similarity between interval time-series data using the left singular vectors of singular value decomposition. We discuss the usefulness of the proposed method using an example of gesture motion.
    Download PDF (569K)
  • Mitsuhiro Tsuji, Toshio Shimokawa, Shigenori Tanaka
    Session ID: TF2-4
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We try to analyze this three-way structure model by applying INDSCAL MDS and INDCLUS Cluster Analysis. We expect that INDSCAL model ( of Multi-Dimensional Analysis ) may show the geometrical interpretation which offers some interesting structure about the characteristics of safety measures. Furthermore we expect that INDCLUS model( of Clusering methods ) may show the detailed geometrical interpretation of this structure.
    Download PDF (553K)
  • MAKOTO FUJIYOSHI, RYUTARO FUKUSHIMA
    Session ID: TF3-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    We introduce to the summary of IPCC (Intergovernmental Panel on Climate Change) 's fourth assessment report (AR4).In this report, they said that "Warming of the climate system is unequivocal. There is very high confidence that the net effect of human activities since 1750 has been one of warming". It is a matter of life and death for us. We expect a speedy solution of the Global warming problem. For this problem, it is important things the reduction of greenhouse gases (GHGs). We will propose the remote fuzzy control system and other support system which is proposed for the effective method for the reduction of GHGs.
    Download PDF (860K)
  • Kazuki Kuribayashi, Seiji Yasunobu
    Session ID: TF3-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    The Multifunctional actuator is an actuator which target angle and compliance slope are configurable. It is difficult for such actuator to perform well when configurations occur several times in a precise moment. On the other hand, a man is able to control their own muscle properly under various environments while able to achieve their aimed moment. This is because a man has a model of their own musculoskeletal system. In this paper, we propose a technique to control a multifunctional actuator properly by incorporating an actuator's model and control knowledge in a system. It's effectiveness is later confirmed by an experiment.
    Download PDF (288K)
  • Maki Endo, Kosuke Tsuruta, Yumi Saito, Hiroshi Nakajima
    Session ID: TF3-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, energy consumption has been added to the performance indices of quality, cost, and delivery in a production line. Besides saving energy, measuring energy consumption can be used for estimation of the equipment condition to realize prognosis, maintenance, and optimization of a production line. In this article, the estimation method of equipment operation condition is proposed. The experiments were conducted to estimate the performance of the proposed method.
    Download PDF (415K)
  • Yasunari Fujimoto, Toru Yamaguchi
    Session ID: TF4-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    This paper shows to develop intelligent power wheelchair that included function adjusting power volume based on human activities. The intelligent power wheelchair has assistance functions to stand up, walk, and control. The assistance functions to walk and control are expected to adjust the power based on user activities. So, we try to analysis the human activity using chaos theory and to adjust assistance based on the activities.
    Download PDF (2328K)
  • Hidetomo Ichihashi, Tatsuya Katada, Makoto Fujiyoshi, Akira Notsu, Kat ...
    Session ID: TF4-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    This paper proposes to improve the performance of the camera-based vehicle detection system by using the classifier based on fuzzy c-means (FCM) clustering. For reducing computation, the semi-hard clustering approach is applied and the hyper-parameters are tuned by particle swarm optimization (PSO). The new system was introduced to an underground parking lot at Ginza, Tokyo in October 2009. The system was also tested at an outdoor (rooftop) parking lot for a period of two months and achieved the detection rate (sensitivity) of 99.6%. The performance clearly surpassed the initial goal of the project.
    Download PDF (424K)
  • Tomohiro Sato, Kohei Nomoto
    Session ID: TF4-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    When a person walks through a town, huge amount of information develops in front of him rapidly. He perceives part of it, and besides, cognize more limited information. This selection of information is an unconscious process, but it determines whether he can understand the way to his goal or he will get lost, because the spatial traveling is understudied using this cognized information. In this presentation, the relation between the local environmental information cognized during walking through a town and the spatial traveling understanding is studied based on experimental data. And the difference of usage of this information between a person who understands the way and a person who does not is shown.
    Download PDF (506K)
  • Keiichi Ogawa, Hiroshi Tsukaguchi, Yoongho Ahn
    Session ID: TG1-1
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    In historical cities, there are a lot of cultural heritage in urbanized areas, and it is necessary to consider disaster mitigation for urban cultural heritage. However, there are many narrow road links in historical cities and it is a weak point of the areas for firefighting in disaster situation. It is considered that the road monitoring system is one of the effective systems for disaster mitigation in such historical areas. In this research, the effects of road monitoring system for cultural heritage disaster mitigation are estimated. The objective area is the central area of Kyoto City, which is a typical historical city in Japan. The location planning of the monitoring cameras is considered based on the location of cultural heritage and the characteristics of road network. Genetic algorithm is applied to solve the problem. It is shown that the effects of road monitoring system become high by the optimization of the location of monitoring cameras using genetic algorithm.
    Download PDF (517K)
  • Toshinobu Oku
    Session ID: TG1-2
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    The diagram of the territory of points distributed in space is a Voronoi diagram. The ordinary Voronoi diagram is clearly determined by the perpendicular bisector, an equal distance from two points. On the other hand, ambiguity is taken in about the equal distance for two points. That is, the ambiguous boundary means the equal distance generally. The ordinary Voronoi diagram is made of in straight lines, but the Voronoi diagram that has ambiguous boundaries is formed by circles or hyperbolas. In this research, the Voronoi diagram with ambiguous boundaries is applied to architectural forms, such as a hemisphere, a cuboid and a free form. It is shown by this method that a new form can be formed.
    Download PDF (829K)
  • Kazuhiko Sasakawa, Akinori Tani, Yuichiro Yamabe
    Session ID: TG1-3
    Published: 2010
    Released on J-STAGE: November 05, 2010
    CONFERENCE PROCEEDINGS FREE ACCESS
    Recently, in order to improve habitability and/or structural safety, control devices based on passive, active, and semi-active control methods are installed to real buildings. The authors have already proposed an intelligent fuzzy optimal and active control system, i.e.; IFOACS, based on intelligent systems, and the effectiveness of proposed system is verified by results of simulations and shaking table tests. In those systems, Active Mass Driver (AMD) system is employed as an activation method of control forces. To perform effective active control, settings of structural characteristics of AMD become important, and optimization methods of AMD have also proposed using Genetic Algorithm (GA) and Immune Algorithm (IA). In this paper, a new evaluation function focused on absolute maximal response acceleration is employed instead of relative maximal response displacement from the foundation and its control effect is discussed.
    Download PDF (1231K)
feedback
Top