Intelligent Community
Member Tadachika OZONO, Tomoharu HASEGAWA, Yuta NAGATA, Satoshi WATANABE, Osamu KAWAMURA, Mariko TAGA, Manabu MATSUYAMA and Toramatsu SHINTANI
KeywordInformation Retrieval, Infromation Recommendation, Machine Learning and Data Mining
Theme Information Retrieval, Infromation Recommendation and Machine Learning

Various useful information and knowledge exist in a community. But these information or knowledge are held by only a few persons, they aren't make use of them fully. The i_com project aims at construction of the system which supports acquisition and sharing of useful knowledge. Recent years, many people study knowledge management: how to use knowledge in a organization.

Research and development are performed about the following system.

  • Research Support System Papits.
  • Support system of analysis of the data in a company.

    In Research Support System Papits, we regard files on each computer as knowledge promote a share of these files on a community, and aim at efficiency of a research survey. We are researching technology of Information Recommendation which recommends the paper conjectured that the user is interested, and Information Retrieval with accuracy higher than the conventional technique. In addition, Papits is equipped with function that collects automatically papers from users, web sites, etc. In order that a user survey the collected papers efficiently, we are researching technology of classifying a paper automatically for fields. We are developing such technology through deveoping Papits.

    About Support system of analysis of the data in a company, this system is developed for the data offered from our collaborated company.

    I think that a user can obtain easily an informational tendency with supporting arrangement of data, and analysis. Moreover, the user may be able to get the feedback for new product development. This system which can create FAQ easily is developed by using the technology of a classification or a similarity judging. By creating FAQ, the quantity of a question to call center can bereduced and it can contribute to the cost reduction of a company.

    As application of the technology which generates new information from the original data, we consider development of a music generation system. This system creates the ad lib performance of a new melody using the information on the ad lib performed until now.

  • Shintani lab. 2003