Video summarization based on user log enhanced link analysis
MULTIMEDIA '03 Proceedings of the eleventh ACM international conference on Multimedia
Improving web search results using affinity graph
Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval
Improving product review search experiences on general search engines
Proceedings of the 11th International Conference on Electronic Commerce
Page importance computation based on Markov processes
Information Retrieval
Hi-index | 0.00 |
Web link analysis has been proved to provide significantenhancement to the precision of web search in practice.Among existing approaches, Kleinberg's HITS andGoogle's PageRank are the two most representativealgorithms that employ explicit hyperlinks structureamong web pages to conduct link analysis, and DirectHitrepresents the other extreme that takes the user's accessfrequency as implicit link to the web page for counting itsimportance. In this paper, we propose a novel linkanalysis algorithm which puts both explicit and implicitlink structures under a unified framework, and show thatHITS and DirectHit are essentially the two extremeinstances of our proposed method. One importantadvantage of our method is its ability to analyze not onlythe hyperlinks between web-pages but also theinteractions between the users and the Web at the sametime. The importance of web-pages and users canreinforce each other to improve the Web link analysis.Compared with traditional HITS and DirectHitalgorithms, our method further improves the searchprecision by 11.8% and 25.3%.