We put forward a new matching strategy with more elaborate and realistic concepts to improve on the past matching strategy based on keyword. In this paper, we set up a concept lattice for each web page according to the natural semantic concepts extracted from each web page. Such concept lattices are encoded by bidirectional associative memories as apposed to complex lattices building algorithms. Then we extract the formal concepts? objects and attributes those are relevant to queries? keywords to carry on matching manipulate.
Citation:
Bo Deng, Yajun Du, "Research on Matching between User Queries and Web Pages Based on BAM," snpd, vol. 3, pp.81-85, Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD 2007), 2007