The increase of the data processed electronically is being enhanced in recent years, and it also becomes possible to get these information easily. However, it is difficult to peruse a lot of data one by one, and to go. Then, an automatic classification of data using machine learning as one of the technology for supporting finding efficiently the information which a user needs is performed. By classifying data into the categories, a user can investigate and analyze intensively the category in which the user is interested.

By using this technology, it is possible to reuse effectively intellectual property, such as know-how and technology in an organization. For example, We have applied to the research support system Papits. Papits is equipped with function that collects automatically papers from users, web sites, etc. By classifying a paper automatically for fields, it is possible for a user to survey user's interested field intensively and easily.

Moreover, application is also considered to the system using the Q&A example in the call center of a company. I think that a user can obtain easily an informational tendency and the feedback of new product development by using this classifing system.