Sentence Similarity Using Syntactic and Semantic Features for Multi-document Summarization

Multi-Document Summarization (MDS) is a process obtaining precise and concise information from a specific set of documents which are on the same topic. The generated summary makes the user to understand the content in a set of documents. The existing approaches suffer with the lack of establishment of semantic and syntactic relationship among the words within a sentence. In this paper, a novel unsupervised MDS framework is proposed by ranking sentences using semantic and syntactic information embedded in the sentences. Empirical evaluations are carried using lexical, syntactic, and semantic features on DUC2002 dataset. The experimental results on DUC2002 prove that the proposed model is comparable with existing systems using various performance measures.