Automatic Summarization Considering Time Series and Thread Structure in Electronic Bulletin Board System for Discussion
2016 5th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI) (2016)
July 10, 2016 to July 14, 2016
On electronic bulletin board systems for discussion, a topic diversifies into multiple subtopics, and the entire structure becomes complicated. It is helpful to show users summarizations of the arguments because they can help in understanding the contents more easily. The purpose of this paper is to propose an automatic summarization method of a single thread considering time series, reply relationships and user information. In the proposed method, all threads are structured in several clusters by hierarchical clustering, and important sentences are selected from each cluster using LexRank, which is a stochastic graph-based method for computing the relative importance of textual units. Finally, we conducted quantitative and qualitative analysis by comparing the proposed method with MMR. The experimental results demonstrate that the proposed method can reduce redundancies more and extract fewer sentences unrelated to the whole context of the summary than MMR. However, the proposed method included fewer important words than MMR.
Time series analysis, Support vector machines, Agriculture, Electronic mail, Redundancy, Context
R. Kitagawa and K. Fujita, "Automatic Summarization Considering Time Series and Thread Structure in Electronic Bulletin Board System for Discussion," 2016 5th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI), Kumamoto, Japan, 2016, pp. 681-686.