文章基本信息

标题：PSEUDO GENETIC AND PROBABILISTIC-BASED FEATURE SELECTION METHOD FOR EXTRACTIVE SINGLE DOCUMENT SUMMARIZATION
本地全文：下载
作者：ALBARAA ABUOBIEDA M. ALI ; NAOMIE SALIM ; RIHAB ELTAYEB AHMED 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2011
卷号：32
期号：1
页码：80-87
出版社：Journal of Theoretical and Applied
摘要：Text features, as a scoring mechanism, are used to identify the key ideas in a given document to be represented in the text summary. Considering all features within same the level of importance may lead to generate a summary with low quality. In this paper, we present a feature selection method using (pseudo) Genetic probabilistic-based Summarization (PGPSum) model for extractive single document summarization. The proposed method, working as features selection mechanism, is used to extract the weights of features from texts. Then, the weights will be used to tune features� scores in order to optimize the summarization process. In this way, important sentences will be selected for representing the document summary. The results show that, our PGPSum model outperformed Ms-Word and Copernic summarizers benchmarks by obtaining a similarity ratio closest to human benchmark summary.
关键词：Summarization; Text Features; Genetic; Probabilistic; Similarity; Sentence Score; Features Weights; Binary Selection