MetadataShow full item record
Researchers all over the world everyday have to search through large amounts of research papers and filter them to choose the right candidate. Ranking these documents is the easy way to mitigate the efforts and time consumed to filter out the unwanted papers. Lexical cohesion is the property of the text that we use to find coherence in the document. This coherence, in turn, can be used to find the semantic relationship between the query words and the documents. We present SDPR: an efficient search framework for ranking research papers. Three significant features characterize this system. First, classifying the words according to their importance in the query. Second, we weight the semantic distance between query and the document using spread activation technique in wordnet graph based ontology. Third, in order to ensure the accuracy of the rank, we weight each document based on the term frequency of query words in them. Results demonstrate a considerable amount of improvement over traditional keyword-based searching algorithm and helps in query disambiguation.