Computational Community Interest and Comments Centric Analysis Ranking
Xiaozhong Liu, Vadim von Brzeski · Illinois Digital Environment for Access to Learning and Scholarship (University of Illinois at Urbana-Champaign) · 2009
Ranking is an important subject in information retrieval, and a variety of techniques and algorithms have been developed to rank the retrieved documents and web pages for a given query. However, ranking is also a challenging task, since it is a dynamic problem, namely a user???s interest toward each query changes from time to time and it is difficult to accurately extract user interest over time. In this paper, we propose an innovative method to extract and weight real time community interested topic for ranking. By generating community interest vector (CIV), we compute the probability score that community interests in specific document or web page in the search results based on daily or past few hours user-oriented data, and use this score for ranking.