Exploiting Semantic Annotations for Entity-based Information Retrieval

In this paper, we propose a new approach to entity-based information retrieval by exploiting semantic annotations of documents. With the increased availability of structured knowledge bases and semantic annotation techniques, we can capture documents and queries at their semantic level to avoid the high semantic ambiguity of terms and to bridge the language barrier between queries and documents. Based on various semantic interpretations, users can refine the queries to match their intents. By exploiting the semantics of entities and their relations in knowledge bases, we propose a novel ranking scheme to address the information needs of users.