Multi-document Summarization using Evolutionary Multi-objective Optimization

Text summarization aims to generate condensed summary from a large set of documents on the same topic. We formulate text summarization task as a multi-objective optimization problem by defining information coverage and diversity as two conflicting objective functions. The result solutions represent summaries that ensure the maximum coverage of the original document and the diversity of the sentences in the summary among each other. The initial experiment using DUC2002 multi-document summarization task dataset and ROUGE evaluation metric shows that the proposed method generates high ROUGE score summaries and is comparable to the state-of-the-art summarization methods.