Extract summarization using Concept-Obtained and Hybrid Parallel Genetic Algorithm

Abstract

This paper proposes a special Chinese automatic summarization method based on Concept-Obtained and Hybrid Parallel Genetic Algorithm. The idea of our approach is to obtain concepts of words using HowNet as tool, and use concepts as feature, not words. We construct conceptual vector space model and use Hybrid Parallel Genetic Algorithm to form summary of documents. The automatic evaluation of document's summaries with n-gram score shows the system's effectiveness and feasibility.