Query based optimal web site clustering using simulated annealing
Wookey Lee, Young‐Kuk Kim, Bok Sik Yoon, Jiang Jin Xi · 2008
Clustering is a viable technique to deal with the scaling issue for the web documents, which has been known for complicated combinatorial optimization problem. It is hard to develop a generally applicable optimal algorithm on the web document clustering and classification for which a simulated annealing algorithm is developed. The web document classification problem is addressed as the problem of best describing match between a web query and a hypothesized web object. The normalized term frequency and inverse document frequency coefficient is used as a measure of the match. Test beds are generated on-line during the search by transforming web sites. As a result, web sites can be clustered optimally in terms of keyword vectors of corresponding web documents.