Abstract | ||
---|---|---|
With the development of folksonomy, social annotations are widely used in various fields. As one of the applications of social annotations, query expansion has been proved meaningful and effective. Considering that user quality influences the relevance between annotations and resources, we infer that it can affect the relevance between expansion terms and the corresponding original queries. Therefore, we believe that annotations used by high quality users are more appropriate to serve as expansion terms. In this paper, we propose a novel algorithm to obtain user quality, and then explore two methods to integrate user quality into the query expansion procedure: one is to filter the resources returned from initial retrieval using user quality; the other is to increase the weight of the expansion terms which are annotations given by high quality users. Our experiments on Del.icio.us dataset show that user quality is helpful for selecting good expansion terms. © Springer-Verlag 2012. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1007/978-3-642-35341-3_35 | AIRS |
Keywords | Field | DocType |
query expansion,social annotations,user quality | Data mining,Web search query,Query expansion,Information retrieval,Computer science,Folksonomy | Conference |
Volume | Issue | ISSN |
7675 LNCS | null | 16113349 |
Citations | PageRank | References |
0 | 0.34 | 16 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Qing Guo | 1 | 3 | 5.12 |
Wenfei Liu | 2 | 16 | 2.58 |
Yuan Lin | 3 | 104 | 16.38 |
Hongfei Lin | 4 | 768 | 122.52 |