Knowledge emergence using Total Environment for Text Data Mining

Recently, various tools for text data mining are provided. A user cannot see data from various viewpoints, because a tool is created for a single task. So, a user has to obtain multiple tools and learn how to use or how to transfer data from one to another. Also, many of those are only used for people who are accustomed to data mining analysis. TETDM (Total Environment for Text Data Mining) project, supported by Japanese Society for Artificial Intelligence, aims at constructing a total environment for combining multiple text mining techniques. General users of TETDM can input text data like news on the Internet or comments in SNS or BBS so that they could obtain new knowledge related to the input text data by using multiple techniques flexibly combining in a single total environment. In this paper, after describing TETDM components, a framework for knowledge emergence using TETDM is described. Though most of data mining systems put their weights on their process, the process from output to interpretation is not considered. Therefore, after the process from output to knowledge emergence is clearly defined, a framework for supporting knowledge creation process is described.