A GIR architecture with semantic-flavored query reformulation

Most geographic queries include references to entities (geographic and non-geographic). Grounding such entities is essential to properly understand the user's information need. As statistical-based query reformulation strategies work at term level, not entity level, they don't use the semantic information given by such entities, which is considerably relevant for the types of queries that should be handled by GIR systems. We motivate the need of a semantic-flavored query reformulation approach for geographic information retrieval systems and describe a GIR architecture where query reformulation focuses on i) grounding entities in the query, ii) selecting a reasoning strategy according to the user information need, and iii) generating a reformulated query containing answers and related entities for a more focused retrieval step. Reformulated queries obtain the answers by accessing a knowledge base.