期刊名称：International Journal of Education and Management Engineering(IJEME)
印刷版ISSN：2305-3623
电子版ISSN：2305-8463
出版年度：2018
卷号：8
期号：1
页码：48-58
DOI：10.5815/ijeme.2018.01.06
出版社：MECS Publisher
摘要：Keyphrase extraction from news web pages is an important task for news documents retrieval and summarization. Keyphrases are like index terms that enclose the important information about document content. Keyphrases actually offer concise and precise description of document content. Key phrases are considered as a single word or a combination of more than one word that represent the important concepts in a text documents. The aim of this paper is to develop and evaluate an automatic keyphrases extraction approach for news web pages. Our approach identifies the candidate keyphrases from documents and chooses those candidate keyphrase having highest weight score. Weight formula combines the feature set that includes TF*IDF, phrase disatnce in documents and lexical chain that is based on WordNet to represent semantic relations between words. The experimental results show that the performance of our approach is better than the contemporary approaches today.
