DESAMC+DocSum: Differential evolution with self-adaptive mutation and crossover parameters for multi-document summarization

Multi-document summarization is used to extract the main ideas of the documents and put them into a short summary. In multi-document summarization, it is important to reduce redundant information in the summaries and extract sentences, which are common to given documents. This paper presents a document summarization model which extracts salient sentences from given documents while reducing redundant information in the summaries and maximizing the summary relevancy. The model is represented as a modified p-median problem. The proposed approach not only expresses sentence-to-sentence relationship, but also expresses summary-to-document and summary-to-subtopics relationships. To solve the optimization problem a new differential evolution algorithm based on self-adaptive mutation and crossover parameters, called DESAMC, is proposed. Experimental studies on DUC benchmark data show the good performance of proposed model and its potential in summarization tasks.