Interactive Independent Topic Analysis for Service

In this paper, we propose a interactive constrained independent topic analysis in text mining. Independent Topic Analysis (ITA) is a method for extracting the independent topics from the document data by using the independent component analysis. In the independent topic analysis, it is possible to extract the most independent topics between each topic. By extracting the independent topic, it is easy to manage the document such as a large number of text data. For example, there are document access support system and document management system. However, the extracted topics by ITA are different from the topic a user requests. In the case of service to the people, interactive system for reflecting the user requests is necessary. We cover the user requests as follows. For example, it is assumed resultant three topics, topic A and topic B and topic C. If a user thought to be close a content of the topic A and topic B, a user wants to merge the topic A and topic B as one of the topic D. In addition, if a user wanted to analyze topic A in more detail, a user would like to separate topic A to topic E and topic F. In that case, it is necessary to incorporate these requests to the ITA. To that end, we define the Merge Link constraints and Separate Link constraints as the user requests. Merge Link constraints is a constraint that merges two topics into one topic. Separate Link constraint is a constraint that separates two topics from one topic. In this paper, we propose a method of extracting a highly independent topic that meets these constraints. We conducted evaluation experiments on proposed methods, and obtained results to show the effectiveness of our approach.