We report on work in progress on using very simple statistics in an unsupervised fashion to re-rank search engine results when review-oriented queries are issued; the goal is to bring opinionated or subjective results to the top of the results list. We find that our proposed technique performs comparably to methods that rely on sophisticated pre-encoded linguistic knowledge, and that both substantially improve the initial results produced by the Yahoo! search engine.
@inproceedings{Pang+Lee:08a, author = {Bo Pang and Lillian Lee}, title = {Using very simple statistics for review search: An exploration}, year = {2008}, pages = {73--76}, booktitle = {Proceedings of COLING: Companion volume: Posters} }
This material is based upon work supported in part by the NSF under grant no. IIS-0329064, a Yahoo! Research Alliance gift, Google Anita Borg Memorial Scholarship funds, a Cornell Provost’s Award for Distinguished Research, and an Alfred P. Sloan Research Fellowship. Any opinions, findings, and conclusions or recommendations expressed are those of the authors and do not necessarily reflect the views or official policies, either expressed or implied, of any sponsoring institutions, the U.S. government, or any other entity.
