A Hybrid PSO Model in Extractive Text Summarizer

The World Wide Web has caused an information explosion. Readers are often drowned in information while starved of knowledge. Readers are bombarded with too many lengthy documents where shorter summarized texts would be preferable. This paper presents a hybrid Harmony Particle Swarm Optimization (PSO) framework in an Extractive Text Summarizer to tackle the information overload problem. Particle Swarm Optimization is a suitable technique for solving complex problems due to its simplicity and fast computational convergence. However, it could be trapped in a local minimal search space in the midst of searching for the optimal solutions. The objective of this research is to investigate whether the proposed hybrid harmony PSO model is capable of condensing original electronic documents into shorter summarized texts more efficiently and accurately than the alternative models. Empirical results show that the proposed hybrid PSO model improves the efficiency and accuracy of composing summarized text.