An Imperialist Competitive Algorithm for Persian Text Segmentation

Abstract

Segmentation has been used in different natural language processing tasks, such as information retrieval and text summarization. In this paper a novel Persian text segmentation algorithm is proposed. Our proposed algorithm applies the imperialist competitive algorithm (ICA) to find the optimal topic boundaries. It is the first time that an evolutionary algorithm applies in Persian text segmentation. The experimental results show that proposed algorithm is more accurate than other Persian text segmentation algorithms.