Name
Abstract | ||
|---|---|---|
Understanding users' search intent expressed through their search queries is crucial to Web search and online adver- tisement. Web query classiflcation (QC) has been widely studied for this purpose. Most previous QC algorithms clas- sify individual queries without considering their context in- formation. However, as exemplifled by the well-known ex- ample on query \jaguar", many Web queries are short and ambiguous, whose real meanings are uncertain without the context information. In this paper, we incorporate context information into the problem of query classiflcation by using conditional random fleld (CRF) models. In our approach, we use neighboring queries and their corresponding clicked URLs (Web pages) in search sessions as the context infor- mation. We perform extensive experiments on real world search logs and validate the efiectiveness and e-ciency of our approach. We show that we can improve the F1 score by 52% as compared to other state-of-the-art baselines. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1145/1571941.1571945 | SIGIR |
Keywords | Field | DocType |
web query classification,query classiflcation,search session,web search,search intent,query classification,search query,individual query,search context,context-aware query classification,context information,real world search log,web query,web pages | Conditional random field,Web search query,F1 score,Data mining,Query language,Web page,Semantic search,Query expansion,Information retrieval,Computer science,Web query classification | Conference |
Citations | PageRank | References |
93 | 2.97 | 27 |
Authors | ||
7 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Huanhuan Cao | 1 | 957 | 33.09 |
| Derek Hao Hu | 2 | 443 | 20.86 |
| Dou Shen | 3 | 1224 | 59.46 |
| Daxin Jiang | 4 | 1316 | 72.60 |
| Jian-Tao Sun | 5 | 1629 | 74.03 |
| Enhong Chen | 6 | 1235 | 86.93 |
| Qiang Yang | 7 | 17039 | 875.69 |