RebaCQ: Query refinement based on consecutive queries

Previous studies reveal that half of the queries submitted to search engines have no follow-up click-through data. This may indicate that users are either dissatisfied with the performance of current search engines or have difficulty formulating correct query keywords related to their search intents. To address this issue, this paper proposes a query refinement mechanism called RebaCQ, which can help users obtain satisfactory pages as soon as possible. By reusing user personal wisdom extracted from their previous consecutive queries, RebaCQ can provide refined result sets closer to user intents. Our experimental results show that result accuracy is significantly increased after adapting RebaCQ.