Seeded text auto-summarization: An experience with simplified statistical and fuzzy ranking algorithm

Abstract

This paper deals with the problem of finding the summary of any text around a given contextual theme. A text paragraph given as input is evaluated and processed according to the SEED (theme/context) provided by the user. Prior to accomplish any computation, all the STOP WORDS are removed from the text. The evaluation and the computation of summary is done based on the ranking of each sentences present in the text. A Membership Function (MF) value is defined for each sentence based on the SEED and the frequency of other words occurring in the sentence. Then the groups of ranked sentences are normalized to form in order to obtain a decreasing MF. Finally, based on the MF value, user can decide an alpha cut to present the resulting summary. It is observed that Seeded auto-summarizing technique in conjunction with the two ways ranking mechanism can improvise a better selectivity in the text and deduce better summaries.