Domain-Based Adaptive T-Rank Search Algorithm

PageRank algorithm is used to re-rank the search results according to relations of Web links and to capture the relative Web pages dependent on particular search query terms. According to the features of web links of structural model, a new adaptive T-Rank algorithm for specific topics was proposed in order to yield more accurate results, deal with a problem of topic-drift exists in links among Web pages and to decrease the re-ranked time after new pages being crawled. Analysis and simulation results show that the improved T-Rank algorithm can efficiently save CPU resources, reduce the calculating time, and better resolve the problem of topic-drift.