Recently, a variety of kinds of digital contents came to be delivered by the development of the Internet. Especially, contents that relate to the entertainment and service using it have spread. It is concerned to the online music distribution service recently. It is because the Internet that doesn't receive the restriction of "Time" and "Space" is used there. However, the number of music that maintained service became huge because there was no space restriction. In addition, it might be difficult to retrieve the music suitable for user's purpose. In this paper, it pays attention to the retrieval that considers person's Kansei (sensibility), and "music" in which the user's preference is strongly reflected is considered. Especially, it pays attention to a difficult retrieval of the classical music in the existing retrieval method, and it approaches from a psychology viewpoint. We executed the questionnaire paying attention to the standard by which the image of music was decided. In addition, we analyzed the principal ingredient, and obtained a useful result.

View full abstract

In recent years, the color variations of the products have been increasing with the diversified human individuality. And also aftermarket services have been active, so we can customize the colorings of the products we bought, for example, a car, a cell phone and so on. So there is the demand for recommendation of suitable color that represents the expression we imagined, and such demand will be increasing. In this paper, by assuming that there is a hierarchy of kansei words, we propose a model of the mechanism that we assign kansei words given physical sensation. In addition, we present a product color coordinate system that reflects user's preference.

View full abstract

In this paper, a new framework for conceptual design support in the field of Computer Aided Design of manufacture is proposed. In the framework, an interactive approach between human and system is introduced, which is named Interactive Genetic Algorithm. Users evaluate pairs of tableware, which are generated as chromosomes by the system, based on the subjective feeling. Then the system re-generates next generation candidates based on the evaluation results. Repeating the interaction, users can acquire pair of tableware which reflects users' concept.

View full abstract

We are studying a robot vision system that has sleep and wake functions. Our proposed mathematical Activation-Input-Modulation (AIM) model, that can express consciousness states such as waking or sleep, controls a ratio between external and internal information processing. The purpose of this paper is to design a new memory architecture like a human's brain for an intelligent real-time robot vision system that can store environmental information in its memories efficiently. The system is equipped with working, short-term and long-term memories. Each memory is divided into explicit and implicit memories. The working memory stores visual information for a few seconds. Information stimulated by external sensors is stored in an explicit short-term memory. Information without a stimulus is stored in an implicit short-term memory periodically. Information organized during sleeping is stored explicit and implicit long-term memories. Operations of these memories, which run in parallel, are also controlled by the AIM model.

View full abstract

We focused on using natural interface for interaction between a human and robots. Recently, robots that fit into the human community are needed. A gestural interface is important for these robots. Actually, gestural recognition has already been studied for use in a human-machine interface. We investigated to find a system that recognizes users' intentions by communication. Therefore, firstly, we constructed a system, which a use point out an object using virtual room interface, our aim is not to calculate accurately with user. Moreover, we show that a robot direct a user's attention to an object using pointing behavior. To acquire a pointing behavior, we constructed virtual robot. We implement the motion, which we acquired by virtual robot to actual robot.

View full abstract

Decision making process of virtual soccer robot has large scale state variable that has 46 independnt parameters. This paper introduces 2-Dimensional arbitrary shaped fuzzy set on the soccer eld to denote linguistic notion associated with soccer strategy, e.g. "right corner", "center middle", "shoot chance area". We show its implementation: OZED system, and show characteristics of them and effciency in rule making.

View full abstract

In this paper, the intention recognition system using the operation recognition because i-mobility set up all over the city centered human understands human operation is proposed. We experimented in intention recognition of suspicious person. This system is loaded into Fuzzy Association memory system and case-based reasoning is executed. And this system makes the case node from target motion and speed. This model memorizes action as a fuzzy set from a case. Arranging a case node performs inductive learning, and the rate of recognition is improved by the increase in a learning example. Finally the automatic addition experiment of a case base is conducted and the convenience of a system is raised. Intention recognition system is explained, and shows its validity of the intention recognition system of operation recognition.

View full abstract

Architectonics of systems and communities are becoming more complicated everyday. Communities help one another by riding on their own strength. Similarly, each matured technology needs help mutually to develop. In particular, the wide development that covers not only 'Robot Technology (RT)', but also 'Automotive Technology (AT)' and 'Information Technology (IT)' is expected in nowadays. They are growing in a spiral by interacting one another. We introduce systems in which these technologies collaborate and our approach for intelligent systems. Finally, we show an experiment of Networked Robots and ontological network.

View full abstract

This paper proposes a framework for acquiring a low-level behavior of a soccer agent. The task of a learning agent is to mimic the behavior of a target agent with a well-trained behavior. Neural networks are used to represent the behavior of the target agent. In order to obtain a set of training data, we convert game logs of the target agent into a set of input-output pairs for the neural networks. We show the effectiveness of the proposed framework through computational experiments.

View full abstract

This paper proposes a TV program recommendation system based on human-robot casual communication under living environment. In these days, partner robots which live together with human are promising for various applications. For the purposes, partner robots do not only have to follow user's every command for prearranged procedures and services, but also communicate with humans in similar ways as human to human communication. In this paper, we propose a human-robot communication system for recommending TV program and related information based on user's interests. The system gathers user's TV watching log and his/her utterances with speech recognition module, based on which his/her interests are estimated and represented as a user profile with the form of bookmarks. A user profile is used for recommending information in which the user is interested. This paper introduces the system architecture and the algorithm for generating user profiles from TV watching logs.

View full abstract

Recently various robots spread in the home, and, in the future, familiar partner robots seem to appear in human life. Therefore, it is necessary that robots learn desirable behavior in order to establish a life of the cooperation between robots and human. We had proposed the methods using interactive GAs so that robots acquire the behavior that the human likes. In addition, we applied this method to the pet robot AIBO. In this paper, we try method, which the robot acquire the behavior, with human interaction using the touch sensor.

View full abstract

Rough set theory was proposed by Z.Pawlak in 1982. This theory can mine knowledge by decision rule from an objective set. The decision rule is used for data analysis well. And we can reason for an unknown object using the decision rule. In this paper, we apply the rough set theory to analysis of economic time series data. We can acquire knowledge from economic time series data and suggest a method to apply to predictions.

View full abstract

This paper discusses the selection support of the leather shoes using the rough set. It is said that one's feet will strongly influence the image. However, the amateur hesitates by the situation which shoes to only have to choose because leather shoes of many kinds are sold in the world now. Then, we pay attention to leather shoes with the many chances to put on official occasions. It especially pays attention to the wedding ceremony in the event. First of all, the image of type was investigated. Afterwards, the annexation rule that is appropriate for the image has been extracted. Leather boots and shoes that is appropriate for the wedding were able to be supported by showing both the rules selecting.

View full abstract

The fashion is a means to express oneself from the young person to the senior citizen, and an important element to enrich our life. Especially, it is also true fashionably in the dress that the number of those interesting of not only the woman but also the man has increased. The designer discovers the characters of a feature color and the shape, etc. paid attention to from the sight image by empirical knowledge so. This is obtained from a lot of experiences of the designer, and, in general, the support system for people is needed. In this paper, we propose a method of extracting the feature of man's suits of "good-looking", and it applies to the design support system of the suit. Moreover, when buying it, those combinations are considered because it puts on the suit at the same time with the shirt and the necktie. For the purpose of our system, we show that the image to the combination of three points of the jacket, the shirt, and the necktie is extracted by using the decision rule of the rough set.

View full abstract

Decision rules with minimal conditions are obtained from a given decision table by a rough set analysis. When the decision table represnts subjective evaluation by a person, such minimal decision rules can be regarded as the essence of his/her opinion. From this point of view, we propose a rough set based apporach to analyze group opinions under given multiple decision tables. In this paper, an agglomerative hierarchical clustering method for multiple decision tables is proposed for the analysis of opinion distribution and several common opinions in the group.

View full abstract

Rough sets are applied to data tables containing possibilistic information. A family of weighted equivalence classes is obtained, in which each equivalence class is accompanied by a possibilistic degree to which it is an actual one. By using the family of weighted equivalence classes we can derive a lower approximation and an upper approximation. The lower approximation and the upper approximation coincide with those obtained from methods of possible worlds. Therefore, the method of weighted equivalence classes is justified.

View full abstract

A flight robot is with wide range move flexibility large since moving in 3-dimensional space is easy. When Blimp robot is compared with other flight robots, prolonged movement is possible for Blimp robot. Furthermore, Blimp robot has high safety to crash, and can use it safely also in indoor space. We developed the Blimp robot hovering control method by present, and realized stable control of a straight line and rotation operation, or operation called vertical movement using the control method. However, in the architecture design of Blimp robot of the conventional research, there was a problem that software creation of autonomous control was difficult. Moreover, Blimp robot tended to be influenced of the air current, and there was a problem that it was difficult to rectify the influence of an air current. In this research, in order to completely autonomous indoor Blimp robot to round indoor automatically was verified..

View full abstract

In this paper, we propose a method of learning docking control of autonomous recharging for indoor blimp robot. Indoor blimp robot can fly in a three dimensional space. For this feature, indoor blimp robot might be applied to entertainment flight in an event hall, and searching for human beings in destructive buildings, etc. To achieve these applications, it might be necessary to fly for long term and to trace orbit accurately. There are many researches for tracing orbit for blimp robots, but it is rare to research for long term flight for indoor blimp robot. Indoor blimp robot cannot load heavy battery for flying longer because of payload restriction. To achieve long term flight, one approach is autonomous recharging. We proposed autonomous recharging system in the past papers. But its control does not be enough to accomplish every time recharging. We designed a new control system for blimp robot. This control is based on learning control. We compare a new approach with a past approach. In the end, we conclude that a new approach is suitable for an autonomous recharging for indoor blimp robot.

View full abstract

We propose the new form of the airship robot: the coupled balloon airship robot. This robot has two or more balloons. The head of each balloon is connected to the other tail. Only one of the balloons can control movement and attitude and tow the other balloons. The advantage of this robot is it has many free payload in the gondolas other than active controllable one. In this paper, we use Memory-based PID controller for the flight control of the coupled balloon robot. This method; one of memory based modeling methods, can adjust the PID parameters in an on-line even if reference signals are changed, and/or system parameters are changed. We describe construction of airship robot hardware, design of flight control architecture and results of experiments in a computer simulation and real world. From the experimental result, effectiveness of a Memory-Based PID method was revealed to control of an airship. In control of the coupled balloons airship robot using Memory-Based PID method, high performance of the control stability better than normal airship robot was indicated.

View full abstract

Recent year, increasing fine showings by robots in several filed. There are research actively, especially autonomous robots which have perceptive senses. Nowadays, autonomous flying robot which can move 3-dimentional space is researched. Flying robots has the advantage that it can operate the area where human or other type robots can not go into because of overhead location or rubble. This paper cover indoor blimp robot. It can keep floating without controlled. Because, it gets buoyancy by filling flotation gases. Therefore, it is able to mobe safty and low energy compared with other flying robot. But indoor blimp robot has some problem such as payload and airstream. And indoor blimp robot can not carry various sensors because low payload ability. And, because this paper focus indoor blimp robot, it can not use GPS, which is most famous positional system. Inertial force and air resistance is greatly affection. Therefore, the time lags occur. So, it is necessary to establish of robust control system to effect of external forces. In this paper, in door blimp robot is navigated by vision-based using potential method and simple image processing algorithm to extract attention area from image.

View full abstract

The ministry of Health, Labor and Welfare of Japan introduced a Long –Term Care Insurance system of April 1st 2000, then started a revised system in April 1st 2003. The eligibility assessment executes classification of patients by tree-based algorithm which enables to classify patients into seven classes depending on the amount of care needed by the eligibility assessment of the Long Term Care Insurance. The problem of finding the necessary conditions for classifying patients into each class is the reverse problem of classification. This article proposes an algorithm for the reverse problem of classification. The computational experiment reveals the effectiveness of the proposed algorithm. Sensitivity analysis is also described.

View full abstract

Sapporo Medical Association plans to develop a regional medical information reference system for supporting patients' migration flow; patients flow from/to various types of medical facilities. This article reports the results of the survey of current Patients' migration in Sapporo city.

View full abstract

Sapporo Medical Association develops a regional medical information reference system for supporting patients' migration flow; patients flow from/to various types of medical facilities. The system consists of three subsystems: the facilities database, the matching subsystem and the coordination subsystem. Facilities database includes medical facilities in Sapporo Medical Association with map information, the matching subsystem solves graph matching algorithm with multi-objectives and the coordination subsystem consults consensus making for both facilities and patients. The trial system starts its running on May 22nd 2006. This article surveys current results of Patients matching system of Sapporo Medical Association.

View full abstract

In recent years, Aging advances and the demand of the public nursing care system has risen in Japan. It is necessary to pass certification of eligibility for long-term care to receive nursing and There are some problems in the recognition system. It aims to pay attention to the first certification of eligibility for long-term care judgment in the public nursing care system in this research based on the background. And to consider the middle evaluation item score becomes the judgment material of the first judgment based Judgment that uses neural net work.

View full abstract

Recently, long-term care insurance is in force in aged society in Japan. Certification of eligibility for long-term care is the need for the using service. The problem is pointed out to the recognition process of first judgment, the possibility that the difference at the nursing time happens after it diverges from the tree structure it is and there is an anxiety of height etc. It proposes the methodology of which it is basic is AHP to solve the problem of originating in such a tree structure in this research.

View full abstract

We have developed a self-assessment system that can be easily used for hotel classification. In this paper, we show the relationships between tourists and hotels in recent years and point out some of the problems in the Japanese hotel market. And we outline that system and explain the effect such a self-assessment system may have on the market in Japan. This is the first trial in Japan of a hotel classification system we have developed in which an expert doesn't evaluate the hotel but the owner or manager evaluates his hotel by himself. This system will contribute to improving not only the relationship between tourists and hotels but also each hotel's management strategy.

View full abstract

Websites are classified into various categories so that it is shown in the directory-based search engine like Yahoo! Japan. Nouns the feature in each category should exist because Website in the category has contents related to the category. If nouns the feature in such each category can be used, websites can be classified into categories. In this paper, it proposes the automatic website classification technique by the similarity with the category based on the score of all nouns extracted in websites, using the appearance frequency of each category. The effectiveness is verified from 10 categories in the directory-based search engine Yahoo! Japan.

View full abstract

The trace system for tourist is proposed in this study. A cellular phone is used in the system. The phone can read QR code and collect the tourist behaviors, namely it can trace the behaviors auto matically. The proposed system was examined in Yamanaka-Onsen in Ishikawa Prefecture, Japan. An electric bulletin board was also adopted to understand directly the tourist opinion. It is considered that the system is useful to know the tourist behaviors and the feeling, and can be contribute to grow the tourism industry.

View full abstract

In this paper, we consider the personalized tourist information system based on AHP~(Analytic Hierarchy Process). This Web based system was consisted by the questionnaire, database of tourist spots and recommendation engine. Each tourist spot in this database is prior characterized by local resident based on questionnaire. Then, it is rated for each tourist by using pair comparison method on demand, the recommended tourist spots or plan are displayed for each tourist. We operated such WWW server based on the proposed approach for Tottori tourism during several years.

View full abstract

In this paper, we propose a support system for tour planning that can be used from PCs and portable devices via http protocol. In this system, when a tourist specifies the starting location, the end location, the usable time and the candidate destinations with relative importance, the departure time and the arrival time, the nearly best schedule is automatically computed. We model tourism as a combinatorial problem and give a practical example of Hokkaido campus adopted as Testbed. In particular, we examine computing time that is the most inevitable problem when this support is operating. We assume that practical computing time for tour planning is ten seconds and compare heuristic and exact algorithms in terms of attraction of total tour. Our experimental results show that if we conduct this comparison and register these results in advance, our tour planning system can compute the nearly best schedule in reasonable time.

View full abstract

We propose an idea to organize photographs stored over the internet. In this study the spatial hyperlink function of Photowalker is extended so that pseudo-3D spaces of photographic images can be seamlessly linked. In addition, we developed a map interface to help the user to create new links.

View full abstract

In this report we propose an intuitive access interface for historical image database. This interface system is build up on a dual map viewer which displays a historical map and a current map simultaneously. We find the spatial correspondences between old and current maps based on landmark location determined from a GIS database. Using this dual viewer we developed an intuitive old photograph retrieval system.

View full abstract

In this paper we propose an integrated system for generating and exhibiting massive-sized digital panoramic photo. This digital photo is generated by amalgamating about 3000 photographs of the city of Hakodate, each taken from different camera posture by a camera control system. A number of image processing technique were applied to form the panoramic photo which size is about 13G Pixel. The photograph is hierarchically structured to enable a pixel-on-demand-based realization of a seamless viewing system.

View full abstract

The technology of taking a picture exists from the end of Edo era. A lot of photographs in which it took a picture of the landmark from the end of Edo era have been left in Hakodate. It is expected that it is possible to experience by comparing Hakodate at that time to present Hakodate in the old photograph those who go sightseeing not are before. However, it is not easy to say that enough under the present situation cooperation will be taken for use as resources for tourism. In this thesis, to use those resources for tourism effectively, the sightseeing information navigation system using GPS and GIS is described. The system acquires sightseeing person's location information from GPS, and offers the image of the old photograph to those who go sightseeing referring to the GIS layer of resources for tourism. The mechanism of the system, the experiment result of the system, and consideration to the result are described at the following.

View full abstract

In this speech, we consider what FUZZY LOGIC is. It is trivial that it is not the usual classical two-valued logic. And, we consider what the logic is.

View full abstract

Developmental patterns of human's various functions are not the same. Therefore, it is necessary to investigate characters of the developmental stages based on developed change in various functions. College student subjects were asked to evaluate what degree 1 year old, 3 years old, 5 years old, 10 years old, 20 years old, 50 years old and 80 years old people would be able to do 20 evaluation items well by themselves. The pattern of developmental change in various functions was different. Methods of the definition and the explanation of the developmental stage by applying the fuzzy set theory are discussed.

View full abstract

In this article we see some relationship between rough sets, modal logic, and topological spaces. In particular we put emphasis on the role of Scott-Montague (neighborhood) models for modal logic and give some comments on the results of Kondo on topologies in generalized rough sets.

View full abstract

This presentation makes some consideration between rough and fuzzy sets from a topological point of view.

View full abstract

This paper shows advantages of Choquet integral type fuzzy switching functions. To close to human thinking, we introduce the rule of representative points instead of the law of excluded middle. Secondly, we examine the meanings of true(1) and false(0) and shows the false has two meanings. Choquet integral type fuzzy switching functions satisfy the rule of representative points and can handle the two meanings.

View full abstract

Many types of fuzzy truth values have been proposed, such as numerical truth values, interval truth values, triangular truth values and trapezoid truth values and so on. Recently, other type of fuzzy truth values (which are called multi-interval truth values) have been proposed and discussed. The characteristic feature of multi-interval truth values is that some of the truth values are not convex. This fact makes the algebraic structures of multi-interval truth values be complex. So, this paper clarifies some of the properties on functions over multi-interval truth values, when the number of variables and the number of non-convex multi-interval truth values are restricted.

View full abstract

Usually, trapezoid truth value (TTV) is investigated in the one-dimensional truth value space V=[0,1]. This paper proposes another viewpoint about the logical operations of TTV. In 1993, T.Oda introduced the Fuzzy-set Concurrent Rating (FCR) method. The two-items type FCR method uses two independent rating scales for a question. Several equations have been proposed to measure the degree of contradiction or irrelevancy in the paired responses. In order to define the indexes for utilizing the paired data obtained by using the FCR method, the two-dimensional Hyper Logic Space (HLS) model is developed. The new model of TTV is derived from the study of the HLS model. The HLS is a kind of extended fuzzy logic. It treats the Truthfulness (T) and the Falseness (F) of the proposition independently. In 1990, Mukaidono and Kikuchi independently treated T and F in their Between Fuzzy Logic (BFL) model. The difference between the HLS and the BFL is, only the definitions of each negation operations. The HLS satisfies the Kleene's law while the BFL does not. In this paper, the outline of the FCR method, the BFL model and the HLS model are also introduced for the understanding of the new TTV model.

View full abstract

A fuzzy (linguistic) truth value is a truth value specified by a fuzzy set over a closed interval $[0,1]$. Logical operations, defined according to the extension principle, do not satisfy all identities to be lattice because there are subnormal truth values and non-convex truth values that violate the absorption laws. In 2000, Brzozowski proposed de Morgan bisemilattice, which is generalized algebra of de Morgan lattice in order for applications in multi-valued simulations of digital circuits. This paper studies a notion of fuzzy-interval equivalnt relation defined for fuzzy functions taking linguistic truth value.

View full abstract

This research work focuses on formulating uninorm based logic (UL) as a weaker logic than monoidal t-norm based logic (MTL). It is shown that the system UL is complete with respect to the class of commutative residuated lattice satisfying prelinearity condition, and that it is an extension of intuitionistic substructural logic FLe.

View full abstract

In this paper, we consider correspondence relationship among belief change, rough sets and information retrieval. First, to consider common parts of these three frameworks, we introduce a situation of information retrieval that a user searches information in a given information table by queries until getting necessary information. We discuss that this situation illustrates correspondence as frameworks among belief change, rough sets and information retrieval. Moreover, we also discuss that operations in these frameworks, that is, operations of information retrieval, operations of belief change, and control of the degree of granularity, correspond each other. This indicates the existence of ``cycles" among these operations.

View full abstract

D.G.Schwartz introduced the formal system of subjective epistemic reasoning in which belief was measured along a series of linguistic degrees, e.g., unequivocally believes, strongly believes, fairly confidently believes, somewhat believes, neither believes nor disbelieves, somewhat disbelieves, fairly confidently disbelieves, strongly disbelieves, unequivocally disbelieves. The purpose of this paper is to present an axiomatization of D.G.Schwartz system, by using the matrix of many-valued logic. Then we prove the completeness theorem as well as the soundness theorem.

View full abstract

This research work treats multiplicative conjunctive operations in type 2 fuzzy logic. In the previous work, the authors gave the axiomatic definition of type 2 t-norms, and proposed a method to construct them from the conventional (i.e. type 1) t-norms via the extension principle, which were named extended t-norms. In the present report, the authors investigate in detail the inclusion relation between the set of type 2 t-norms and that of extended t-norms, and also give the necessary and sufficient condition for type 2 t-norms so that they could be expressed as the extended t-norms.

View full abstract

It is well known that the extension principle plays an important and fundamental role in fuzzy analysis. In this paper, we have obtained the condition that a real function transforms a fuzzy function using the extension principle. And we discuss some properties of the exponential function on the Fuzzy Number Space.

View full abstract

It has been already known that a set function $\mu$ defined on the power set $2^X$ of a finite set $X$ and/or the Choquet integral w.r.t. $\mu$ have a $\mathcal C$-decomposition iff $\mathcal C$ is an inclusion-exclusion covering of $X$. This paper shows that for any two positive integers $k$ and $k'$ such that $k'\le k$, there exist a finite set $X$, a $k$-monotone set function $\mu$ on $2^X$, and an inclusion-exclusion covering $\mathcal C$ of $X$ w.r.t. $\mu$ such that $\mu$ has no $k'$-monotone $\mathcal C$-decompositions.

View full abstract

In other research papers, we proposed the distance type fuzzy reasoning method based on distance information between fuzzy sets. In this report, first of all, to achieve human reasoning functions, three important concepts are defined about the reverse- reasoning system. Finally, we will discuss the characteristic between the reasoning units that compose by using the distance type fuzzy reasoning method.

View full abstract

We introduce a parametric representation of fuzzy numbers with bounded supports as well as we consider a normed space including the set of fuzzy numbers, where the addition in the normed space is the same one due to the extension principle but the difference and scalar products are not the same as those of the principle. In this article we treat the Frechet differential in a Banach space of fuzzy numbers and we discuss variational equations of fuzzy differential equations in order to get improved results on the stability analysis of fuzzy differential equations.

View full abstract