Extractive Summarization Of Farsi Documents Based On PSO Clustering

As there is an ever-increasing number of textual resources, users nowadays enjoy access to a wider range of data; hence, accessing accurate and reliable ones has become a problematic issue. Automated summarization systems can play a principal role in covering the main ideas of the texts and removing time limitations. The present study presents a textual summarization system based on sentence clustering. There are some methods proposed to solve clustering problems so that for reaching a desirable clustering, collective intelligence algorithms are used for optimizing the methods. These methods rely on semantic aspect of words based on their relations in the text. Ultimately, appropriate sentences are selected from each cluster after clustering the sentences on the basis of the aforementioned criteria. A collection of Persian sports news articles are selected for the assessment. The findings reveal that the presented method yields more accurate results than others.