Extractive single-document summarization using adaptive binary constrained multi-objective differential evaluation

The incredible growth of the Internet has enhanced the research and development of automatic text summarization. Several approaches were proposed in the literature for automatic text summarization using a Multi-objective optimization (MOO) framework. But, still, it is difficult to decide which feature set and objective functions are best suited for the summarization task. Improving these objective functions along with the suitable optimization framework can bring diversity among the solutions and convergence towards genuine Pareto optimal fronts. So, it can be fascinating to prospect other proficient techniques that can further improve the performance of the automatic summarization process. This work proposes an adaptive binary constrained differential evolution (ABCDE) technique in the MOO framework for solving the summarization problem. The implemented system significantly outperformed various existing methods on ROUGE measures when evaluated on DUC 2001 and DUC 2002 data sets. Obtained results illustrate the supremacy of the proposed approach in terms of ROUGE scores, readability, and relevancy.