The Web of Data is growing in popularity and dimension,udand named entity exploitation is gaining importance in many researchudfields. In this paper, we explore the use of entities that can be extractedudfrom a query log to enhance query recommendation. In particular, weudextend a state-of-the-art recommendation algorithm to take into accountudthe semantic information associated with submitted queries. Our noveludmethod generates highly related and diversified suggestions that we as-udsess by means of a new evaluation technique. The manually annotateduddataset used for performance comparisons has been made available toudthe research community to favor the repeatability of experiments.
展开▼
