Organizing search results is one of the challenging task of the search engines due to various and dynamic intentions of the queries. As a consequence search engines are not able to understand the exact user context, and thus retrieve large volumes of results, most of which are irrelevant to the user. Search Result Clustering (SRC) is a technique which groups the search results and presents users the various intentions of the query. In this work, we have proposed an approach that first identifies the associated topics and represents them in the form of concepts and then forms groups of documents by assigning each document to the appropriate topic and in the end it provides suitable labels to these topics. Experimental results show that the proposed method is able to produce encouraging results as compared to the most popular non-commercial methods Lingo and STC on standard datasets such as ODP and Ambient datasets.
[1]
Dawid Weiss,et al.
Lingo: Search Results Clustering Algorithm Based on Singular Value Decomposition
,
2004,
Intelligent Information Systems.
[2]
Marti A. Hearst,et al.
Reexamining the cluster hypothesis: scatter/gather on retrieval results
,
1996,
SIGIR '96.
[3]
Poonam Goyal,et al.
Concept Based Query Recommendation
,
2011,
AusDM.
[4]
Oren Etzioni,et al.
Web document clustering: a feasibility demonstration
,
1998,
SIGIR '98.
[5]
Dawid Weiss,et al.
A survey of Web clustering engines
,
2009,
CSUR.