Research on Automatic Text Summarization Based on Extractive Method

Abstract

With the rapid growth of text data, automatic text summarization can save a lot of time and energy to quickly obtain the important information of text. Aiming at the application of text feature representation and different abstracts extraction methods, this paper summarizes the common techniques and characteristics of summarization from three stages: text vector representation, summary method selection and sentence extraction. According to the inevitable problems of semantic logic and redundancy in extractive summarization, this paper summarizes the existing methods to deal with redundancy. Finally, the common evaluation methods are introduced.