Querying time-series document clusters (データベースシステム)

概要

論文の詳細を見る
Document clustering methods for time-series documents produce a sequence of snapshots of clustering results over time. Tracking topic evolutions in a large stream of clusters is very important in the analysis of the characteristics of the data and the real world events. For example, for the topic US presidential election campaign, does the topic "Clinton nomination campaign" dissolve and change to "Obama nomination campaign"? Or does "Clinton nomination campaign" merge with "Obama nomination campaign" and form a broader topic "Democrat US presidential campaign"? Analyzing the contents (topics) and trends in a long sequence of clustering snapshots, however, is hard and requires efforts since there are too many number of clusters; a user may need to access every cluster or read every document contained in each cluster. In this paper, we propose a framework to query topic evolution in a large stream of clusters. Specifically, we propose a query language for time-series document clustering results and an approach to query processing.
社団法人情報処理学会の論文
2008-09-14