Approximate Query Results Ranking for E-commerce Web Databases

摘要

To deal with the problem of many-answers returned from an E-commerce Web database in response to an approximate query, this paper proposes a novel approach to rank the approximate query results. Based on the database workload, we speculate how much the user cares about each attribute and assign a corresponding weight to it. Then, based on the similarity of different attribute values, a satisfaction degree ranking method, which ranks the answer tuples according to their satisfaction degree to the initial query, is presented. Next, for the tuples with the same satisfaction degree, each unspecified attribute value is assigned a score according to its "desirableness" to the user, which is used for distinguishing these tuples. Finally, the results of preliminary experiment, which demonstrate the effectiveness of the ranking algorithm, are presented.