Simrank++: Query rewriting through link analysis of the click graph Export

@techreport{citeulike:2075319, abstract = {We focus on the problem of query rewriting for sponsored search. We base rewrites on a historical click graph that records the ads that have been clicked on in response to past user queries. Given a query q, we first consider Simrank as a way to identify queries similar to q, i.e., queries whose ads a user may be interested in. We argue that Simrank fails to properly identify query similarities in our application, and we present two enhanced version of Simrank: one that exploits weights on click graph edges and another that exploits ``evidence.'' We experimentally evaluate our new schemes against Simrank, using actual click graphs and queries form Yahoo!, and using a variety of metrics. Our results show that the enhanced methods can yield more and better query rewrites.}, author = {Antonellis, Ioannis and Garcia-Molina, Hector and Chang, Chi-Chao}, citeulike-article-id = {2075319}, citeulike-linkout-0 = {http://dbpubs.stanford.edu:8090/pub/2007-32}, day = {29}, institution = {Stanford}, keywords = {ads, analysis, behavior, clickstream, graph, information, ir, linking, query, ranking, retrieval, search, user}, month = {October}, posted-at = {2007-12-07 22:26:17}, priority = {2}, title = {Simrank++: Query rewriting through link analysis of the click graph}, url = {http://dbpubs.stanford.edu:8090/pub/2007-32}, year = {2007} }

TY - RPRT ID - citeulike:2075319 L3 - citeulike-article-id:2075319 N2 - We focus on the problem of query rewriting for sponsored search. We base rewrites on a historical click graph that records the ads that have been clicked on in response to past user queries. Given a query q, we first consider Simrank as a way to identify queries similar to q, i.e., queries whose ads a user may be interested in. We argue that Simrank fails to properly identify query similarities in our application, and we present two enhanced version of Simrank: one that exploits weights on click graph edges and another that exploits ``evidence.'' We experimentally evaluate our new schemes against Simrank, using actual click graphs and queries form Yahoo!, and using a variety of metrics. Our results show that the enhanced methods can yield more and better query rewrites. TI - Simrank++: Query rewriting through link analysis of the click graph KW - ads KW - analysis KW - behavior KW - clickstream KW - graph KW - information KW - ir KW - linking KW - query KW - ranking KW - retrieval KW - search KW - user AU - Antonellis, Ioannis AU - Garcia-Molina, Hector AU - Chang, Chi-Chao PY - 2007/10/29/ UR - http://dbpubs.stanford.edu:8090/pub/2007-32 ER -