Extraction and reorganization of topic transition patterns in recorded document

点击翻译按钮获取中文摘要

This research proposed a system which focused chronological order of documents had accumulated over a period of time. It extracted and reorganized topics and its transitions from documents to acquire knowledge. It decomposed each document into smaller text fragments and grouped them into clusters of fragments. We regarded a cluster of text fragments as a topic. The system made document groups based on elapsed time, and calculated relevance of topics in neighboring document groups. Then it obtained topics and their transition. It could sort out topics temporal topics and constant topics. It also reorganized topics and their transition with keywords indicated by a user.