Keyphrase Extraction Using PageRank and Word Features

Abstract

Keyphrase extraction is a fundamental task in natural language processing. Its purpose is to generate a set of keyphrases representing the main idea of the input document. Keyphrase extraction can be used in several applications such as recommendation systems, plagiarism checking, text summarization, and text retrieval. In this paper, we propose an approach using PageRank and word features to compute keyphrases’ scores. Experimental results on SemEval 2010 dataset show that our method provides promising results compared to existing works in this field.