This paper proposes a method to support interpretations of difference between two keyword relationships with varying viewpoints in a map animation interface. Much research has been conducted on extracting keyword relationships using hit counts obtained by search engines. Since relationships also exist in the number of viewpoints, we need to understand the differences among relationships that correspond to each viewpoint. Finding keywords whose relationships drastically change is crucial because the value of keywords mainly consists in their relationships in networks. Therefore, this method marks keywords that drastically change their relations when animations between two relationships are shown in the interface. Experimental results showed that the interface using the method supported difference interpretation better than a comparative interface, which outputs no marks or outputs marks randomly.

View full abstract

This paper proposes a system for supporting discussion about community anticrime actions using bulletin board system (BBS) with view. In recent years, many crimes targeting children have been occurring throughout the country, and the importance of community anticrime actions including anticrime patrols and safety map creation has increased. However, such community anticrime actions have been performed independently without collaboration with others. The system proposed in the paper aims at supporting community anticrime actions comprehensively, including community map creation by children, map sharing, and online discussion about anticrime actions. In order to support online discussion by utilizing the community safety map created by children, the concept of BBS with view is employed. Recent trend in information visualization is to use views in communication and social data analysis, and discussion around views is expected to lead sufficient exploration and deep understanding of data set. In particular, BBS with view focuses on task-oriented discussion, and proposes graphical annotation and message retrieval functions as online discussion support functions. In addition to those functions, the paper also proposes an annotation retrieval function for performing task-oriented discussion more efficiently. The proposed system was actually used in the class of an elementary school, and online discussion based on the obtained map was performed with using the system. The results of analyzing the discussion show the discussion support functions of the proposed system are useful for having concrete discussion while referring to specific spots. Although the support facility of the proposed system is limited, more practical system can be developed based on the system by introducing additional features, such as sharing of criminal and accident information, and support function for avoiding oversight of information during discussion.

View full abstract

GA searches neighborhood of the genes that have high evaluation values as it is assumed that neighboring relationship of individuals in the genotype space is similar to that in the evaluation space. However, it is difficult to decide which coding, deciding the similarity between them, and which genetic operation to use because there is no criterion for the similarity and the landscape of search space. As a result, they are decided by trial and error in a lot of cases. This paper proposes the visualization method to grasp the relationship between genes and their evaluation values for efficient search. This paper applies the proposed method to a benchmark function of multi-objective optimization problem and shows that it enables us to grasp the similarity of genes between in the genotype space and evaluation space. It also shows that we could feed back the visualization result to genetic operations for more efficient search.

View full abstract

An existing financial analysis system doesn't consider user's characteristic, and not be a system that considered the interaction between systems as the user though there are a lot of researches that do the risk and the analysis of the affairs of a business on the enterprise in the financial analysis. There is an analysis angle of financial analysis angle in the financial analysis, and the financial analysis is done generally based on it. The one from here from everything are in information that can be read there even if the financial data that the user demands is decided various the one, and obtained by noticing the person who does the financial analysis though the financial analysis seems to have a specific purpose from the beginning. That is, it can be said that it has the side of exploratory data analysis. The user has it in exploratory data analysis through the search process for this case, the user will have it a lot of actualizing an own information request gradually by repeating the interaction and make to exquisite as for the possibility that the information demand is understood an own information request, and changes gradually and changes in quality during the search. Then, the system that supports the user interaction in the financial analysis support system and the financial analysis support system of the Web base using XBRL is constructed in the present study. The interaction design named Model+Others was proposed, and mounted according to three financial analysis angles. These Model+Others related to the importance in the financial analysis angle, divided a financial index from the relation of the index into three hierarchies, assumed “Model”, and located other parts to “Others”. It is thought that the interaction support is enabled by applying these models. It is time when did not apply in case of occasioned to apply the user model, and user's use log was pursued and analyzed from an ethno viewpoint in detail the verification your what influence being exerted on the use user. The interaction support was done by having used proposed “Model+Others”, and as a result, it was noticed the change in user's demand information, and was able to observe each user's feature interaction result.

View full abstract

A Fuzzy Constraint Satisfaction Problem (FCSP) is an extension of the classical CSP, a powerful tool for modeling various problems based on constraints among variables. Meanwhile, a Dynamic CSP (DCSP) is a framework for modeling the transformation of problems, and the differences between past solutions and new solutions should be as small as possible. These schemes are the techniques to formulate real world problems as CSPs more easily. The CSP model that combines these (Dynamic Fuzzy CSP) has already existing researches. However, as CSPs are NP-hard problems in general, no efficient and complete algorithms for solving CSPs exist and the increase in the worst-case computation time is exponential in the size of the problems. In the work reported in this paper we tested a hybrid approximate method, called the SRS algorithm. Moreover, we introduce a post-filtering method, called the SRSD algorithm. We empirically show that SRS and SRSD algorithms keep the stability of solutions better than other algorithms. They are able to quickly get good-quality approximate and stabile solutions to large problems.

View full abstract

Neurophysiology has revealed the existence of mirror neurons in brains of macaque monkeys and they shows similar activities during executing and observation of goal directed movements performed by self and other. The concept of the mirror neurons/systems is very interesting and suggests that behavior acquisition and recognition of observed behavior are closely related to each other. That is, the behavior learning modules might be used not only for behavior acquisition/execution but also for the understanding of the behavior/intention of other. We propose a novel method not only to learn and execute a variety of behavior but also to understand behavior of others supposing that the observer has already acquired the utilities (state values in reinforcement learning scheme) of all kinds of behavior the observed agent can do. The method does not need a precise world model or coordination transformation system to deal with view difference caused by different viewpoints. This paper shows that an observer can understand/recognize a behavior of other not by precise object trajectory in allocentric/egocentric coordinate space but by estimated utility transition during the observed behavior.

View full abstract

This paper proposes a novel radial basis function network (RBFN) for chemical taste sensor and a method for prediction of optimal substance-concentration of taste. The taste sensor consists of an array of electrochemical sensors and neural networks for gustatory simulation. Our goal is the implementation of a neural-network-based gustatory system that employs the following two functions: (1) estimation of taste from electrochemical sensor responses; (2) prediction of optimal substance-concentration of taste. To implement the former function, we propose a modified version of two-phase optimized RBFN, which has previously proposed by authors, by the adjustment of widths of kernel functions with Parzen window. Furthermore, a new network inversion is proposed for the implementation of the latter function. In the proposed network inversion, the finding rules of the initial input vectors are optimized for modified RBFN. From the experimental results, the estimation accuracy of the proposed RBFN obtained 5.6% higher in average than that of the conventional method. Moreover, the prediction accuracy of the proposed network inversion obtained 10.7% higher in average than that of the conventional method.

View full abstract

In this paper we consider autonomous control of a real snake-like robot using reinforcement learning. We focus on curse of dimensionality and lack of generality, and point out that the causes of these problems are not in learning algorithm but in neglect of properties of the real world. To solve these problems we propose new framework in which the body of the robot abstracts general meanings by interacting with environment. We apply the proposed framework to control of a real snake-like robot and confirm that the two problems are solved.

View full abstract

The present paper considers a multihop teleoperation by mobile robotic sensor network in an unknown environment. A mobile robotic sensor network is composed of a distributed collection of mobile robots, each of which has sensing, computation, communication, and locomotion capabilities. Such a network is capable of changing the network topology in order to gather distributed information to the sink node in an unknown and dynamic environment. To realize this network, we propose a self-deployment algorithm by which robots can explore an environment while maintaining their communication link. In addition, a routing algorithm is developed so that observed data from a robot is transferred to the sink. Furthermore, a mobile robot system inter-linked with peer-to-peer communication is developed. Finally, a multihop teleoperation system is developed for a multi-robot cooperation platform.

View full abstract