A materials knowledge base has been developed with structured data format. The materials theory is represented as conceptual frameworks and illustrated by concept maps. A XML file is used to describe the content of each concept, and a RDF file is used to describe the relationships of the concepts. The knowledge data is dynamically retrieved, organized and converted into web pages including text, graphs, mathematical expressions etc. by a computer. The web pages are linked to some external data and program sources through interactive user interfaces to enrich the presentation of the knowledge.

View full abstract

　The Materials Information Technology Station (MITS) of the National Institute for Materials Science (NIMS), has developed a system for retrieving information from its material property databases. The MITS system provides eleven databases : four concerning structural materials, four databases containing physical properties; two databases concerned with superconducting materials, and one database with considerable information about polymers. These have been available on its website (http://mits.nims.go.jp/) since April, 2003. Granta Design Ltd has independently established the Materials Database Network (http://matdata.net/) by linking materials information from a number of key reference sources, world-wide. These include ASM International's Online Material Information System, and sources of information for aerospace alloys, polymers, materials standards, joining etc. The NIMS and Granta systems have now been combined into a single integrated software environment, which enables rapid-searching of the available reference sources and retrieval of top quality materials information from each of them.

View full abstract

　Huge materials information are available on Web system of the Internet and/or in-house or personal computer systems. The information consist of numerical data, textual data, and image data. Moreover, if additional processings of such data using software tools are carried out, new information are added on the raw data due to the data processings. So as to meet users' needs properly, semantics of each datum along the data processing are to be taken into account in each retrieval service. As the first trial of realizing such proper services, user interface to obtain necessary by taking advantage of the in-house ontology developed by the authors were demonstrated by using three high technology cases, namely creep property in heat-resistant materials, welding as a kind of manufacturing processes and isotope ratio dependencies on intrinsic properties which requires more microscopic descriptions than atomic element composition of material.
　The retrieval for information on the internet, in-house and personal Web were tried using the user-interfaces based on keyword of terminology induced automatically from created four dictionaries in materials field. In the results, it was suggested that new retrieval service is obtained using developed user-interface and data analysis tool, if the retrieval is according to the conceptual map of terminology in three high technology cases.

View full abstract

　We developed the system that organically integrates various distributed databases on the internet for material design and showed its function. This system carries out the integrated retrieval using virtual database which classified and defined search condition item and search result item of each database. This system can search multiple databases simultaneously, and display their search result in the list. Therefore, we expect that this system is useful for preliminary search and overlooking data.

View full abstract

　Databases available in various chemical information areas, i.e., references, chemical substances, reactions, physical properties and spectra, are discussed.

View full abstract

[in Japanese]

View full abstract

　This paper proposes a method to extract multiple target image areas similar to a given template image. The proposed method based on a genetic algorithm (GA) with immune system searches efficiently multiple local solutions by using suppressor T cell's mechanism and local search, which are major improvements in it, whereas genetic algorithms search only one global solution. Experimental results show that the suppressive mechanism keeps efficiently diversity of individuals and extracts multiple target image areas in the condition of free locations, inclinations and scales at a high probability.

View full abstract

　It is important to develop information systems which are capable of recommending user information or items reflecting user's individual tastes. We propose a music artist recommendation system that provides user with music artists who are the candidates being fit with user's tastes, and are derived from inference based on "favorite music artists" and "reasons to prefer them". Our system consists of an interface for recommendation and a database storing association rules that are patterns of relations among music artists who are liked by people concurrently. The characteristic of our system is to consider individual differences of reasons to prefer music artists in the recommendation processes. The performance of our system shows that reasons to prefer music artists have an effect on ranking recommended music artists, and that our system is superior to other systems in the recommendation accuracies and computational time.

View full abstract

　Digital information media has become an important information source in modern society. However, a digital divide has been created between people who can use information media and those who cannot. The volume of information available to blind people in particular, who view Web pages using a voice browser and whose use of Web pages relies on audible cues, is significantly less than that available to sighted people.
　We configured a Web Viewing Assistance System for blind people in this study by utilizing the Digital Talking Book Method. This system converts target Web pages into DTBs to provide their contents to users. Operation of the DTB method is uniform and simple for blind people and can provide an operational environment with reduced complexity compared with viewing Web pages using a voice browser.
　In the implemented DTB system, the number of key strokes required to reach target contents is considerably reduced in comparison with that of a voice Web browser.

View full abstract

　We develop the system which automatically carries out Go-record from the image of the Go-game situation. For the Go-record, Go-stone has to decide its position on the 19x19 cells of the Go-board which we call Go-coordinates. In our system, the Go-board is shot in various directions with the CCD camera for simple operation. In this case, it is difficult to directly extract accurate Go-coordinates in the image. However, we are able to extract accurate Go-coordinates if the area of the Go-board is recognized accurately. Consequently, in the first step, the contour of the Go-board has to be recognized from the image which includes unrelated objects. In this paper, we propose the pattern matching algorithm for the extraction of the accurate Go-coordinates in the image using the genetic algorithm (GA). We evaluated the proposed algorithm on the prototype using the proposed algorithm by dealing with the image which includes unrelated objects. As a result, it has been verified that the prototype generates the correct Go-record.

View full abstract

　The aim of our study is to develop a creativity support system which uses analogical reasoning. We adopt patent documents for extracting base knowledge in analogical reasoning on the system because patent documents have the following features; 1) Patent documents have potential to be extracted a sort of causal relationship between a method of invention and its effect, since the relationship is described in patent documents; 2) Extracted causal relationships have potential to be applied to industry, since patent documents include various industrial fields. Therefore we tried to extract some causal relationships from patent documents in fiber engineering, and some causal relationships between compound and its property are obtained. After that we extracted some causal relationships from patent documents in pharmaceutical field, and contrived a method for analogical reasoning by using the obtained causal relationships and similarity of keywords. Those results of our study are described in this paper.

View full abstract