Extracting an Optimal Set of Linguistic Summaries using Genetic Algorithm Combined with Greedy Strategy

Abstract

The goal of extracting linguistic data summaries is to produce summary sentences expressed in natural language which represent knowledge hidden in numerical dataset. At the most general level, human users can get a very large number of linguistic summaries. In this paper, we propose a model of genetic algorithm combined with greedy strategy to extract an optimal set of linguistic summaries based on the evaluation measures of goodness and diversity of the set of linguistic summaries. The experimental results on creep dataset have demonstrated the outperformance of the proposed model of genetic algorithm combined with greedy strategy in comparison with the existing genetic algorithm models in extracting linguistic summaries from data.