Authors: Yi Chen, Yi Shan, Yichuang Cai, Ziyang Liu
Tags: 2015, conceptual modeling
This paper addresses an open problem of keyword search in XML trees: given relevant matches to keywords, how to compose query results properly so that they can be effectively ranked and easily understood by users. The approaches adopted in the literature are oblivious to user search intention, making ranking schemes ineffective on such results. Intuitively, each query has a search target and each result should contain exactly one instance of the search target along with its evidence about its relevance to the query. In this paper, we design algorithms that compose atomic and intact query results driven by users’ search targets. To infer search targets, we analyze return specifications in the query, the modifying relationship among keyword matches and the entities involved in the search. Experimental evaluationsvalidate the effectiveness and efficiency of our approach.Read the full paper here: https://link.springer.com/chapter/10.1007/978-3-319-25264-3_33
