An Efficient Approach For Semantically-Enhanced Document Clustering By Using Wikipedia Link Structure

Abstract

Traditional techniques of document clustering do not consider the semantic relationships between words when assigning documents to clusters. For instance, if two documents talking about the same topic do that using different words (which may be synonyms or semantically associated), these techniques may assign documents to different clusters. Previous research has approached this problem by enriching the document representation with the background knowledge in an ontology. This paper presents a new approach to enhance document clustering by exploiting the semantic knowledge contained in Wikipedia. We first map terms within documents to their corresponding Wikipedia concepts. Then, similarity between each pair of terms is calculated by using the Wikipedia's link structure. The document’s vector representation is then adjusted so that terms that are semantically related gain more weight. Our approach differs from related efforts in two aspects: first, unlink others who built their own methods of measuring similarity through the Wikipedia categories; our approach uses a similarity measure that is modelled after the Normalized Google Distance which is a well-known and low-cost method of measuring term similarity. Second, it is more time efficient as it applies an algorithm for phrase extraction from documents prior to matching terms with Wikipedia. Our approach was evaluated by being compared with different methods from the state of the art on two different datasets. Empirical results showed that our approach improved the clustering results as compared to other approaches.