Fu Xianghua,Ma Zhaofeng,He Ming，Feng Boqin
(School of Electronics and Information Engineering,Xi'an Jiaotong University,Xi'an 710049,China)
Abstract:To interpret the procedure of hypertext induced topic search (HITS) based on a semantic relation model, the reason about the topic drift of HITS was found that Web pages are projected to a wrong latent semantic basis. A new algorithm for personalized topic distillation and hierarchical exploration (PTDHE) was presented to improve the quality of topic distillation. Personalized root set and base set with query expansion was constructed using individual query logs to avoid the topic draft, and applying a hierarchical division algorithm based on min-max principle to explore relative topics of user query, and then using HITS to evaluate and return authority pages of relative topics to end-users.The experimental results on 14 queries show that PTDHE performs better than HITS in topic distillation quality and topic exploration ability. PTDHE reduces topic drift rate by 2% to 66% compared to that of HITS, and discovers several relative topics to queries that have multiple meanings.
Keywords:link analysis;hypertext induced topic search;topic distillation;topic drift; query expansion