A Study on the Use of Genetic Programming for Automatic Text Summarization

Abstract

Text Summarization is the process of identifying and extracting the most vital information in a document. It has been seen as an effective method for dealing with increasing amount of information on the Internet nowadays. In this paper, we present an application of Genetic Programming to the problem of Automatic Text Summarization. Genetic Programming was used to evolve the function that ranks the sentences in a document based on their importance. The summary was extracted by selecting the sentences that have the highest rankings. The experiment was conducted on a number of Vietnamese news documents. The result showed that the summaries created by Genetic Programming are better than those created by a number of statistic based methods and even by human (non-experts).