Topic Based Automatic Text Summarization using Association Rule and PLSI

Abstract—Automatic text summarization plays an important role in information retrieval that is core many tools like search engines, question-answering systems and etc. Automatic text summarization can be extract salient feature from documents, which helps user to get useful information in short time and less effort. In this paper we proposed method for topic based automatic text summarization with association rule (AR) and probabilistic latent semantic indexing (PLSI). This approach to extract topics form a document used of AR that this topics known as concepts which use to identify the most important sentences in a document. Also the PLSI has been used to sentence ranking based on identified topics, that this tool is useful to find the underlying probabilistic relationships between terms and documents. Proposed method were evaluated using ROUGE metrics and evaluation results obtained for DUC 2002 show that our proposed method could improve the summarization results significantly.

Keywords:

Text summarization , association rule , probabilistic latent semantic indexing