Extractive text summarization using Fuzzy-c-means clustering

Abstract

Keeping track of the precise information from a large volume of text is an arduous task for human. Test summarization process has become one of the significant research areas for years owing to cope up with the astounding increase of virtual textual material. Text summarization is the process to keep the relevant important information of the original text in a shorter version with the main ideas of the original text for understanding innumerable volumes of information easily within a short period of time. There are two main classifications of text summarization process, Extractive and Abstractive text summarization. Extractive summarization processes by using most important fragments of exiting words, phrases or sentences from the original document. It largely depends on sentence-extraction techniques or sentence-based model. A sentence based model using Fuzzy C-Means clustering has been proposed this research. Six key features including a new feature have been added for the sentence scoring. Performance of the proposed FCM model is evaluated by ROUGE, which has been gauged with the precision, recall and f-measure.The result shows that this FCM model interprets extractive text summarization methods with a less summary redundancy and depth of information and also it shows more adhering and coherent than other previous approaches. Keywords: Sentence Extraction, Clustering, Summarization.