vioft2nntf2t|tblJournal|Abstract_paper|0xf4ff973519000000076f030001000400 Query expansion (QE) is the process of reformulating a query to improve retrieval performance. Most of the times user's query contains ambiguous terms which adds relevant as well as irrelevant terms to the query after applying current query expansion methods. This results in to low precision. The existing query expansion techniques do not consider the context of the ambiguous words present in the user's query. This paper presents a method for resolving the correct sense of the ambiguous terms present in the query by determining the similarity of the ambiguous term with the other terms in the query and then assigning weights to the similarity. The weights to the similarity measures of the terms are assigned on the basis of decreasing order of distance to the ambiguous term. An aggregated similarity score based on the assigned weights is calculated using Ordered weighted averaging operator (OWA) for each sense of the ambiguous term and the sense having highest similarity score will be considered as the most appropriate sense for the ambiguous term. After then, the query is expanded by taking an implicit feedback from user and adding terms related to the corresponding sense and hence optimizing the query expansion process.
Kanika Mittal1, Amita Jain2
Bhagwan Parshuram Institute of Technology, India1, Ambedkar Institute of Advanced Communication Technologies & Research, India2
Query Expansion, Natural Language Processing, Information Retrieval, Word Sense Disambiguation, WordNet, OWA Operator
| January | February | March | April | May | June | July | August | September | October | November | December |
| 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 5 , Issue: 2 , Pages: 896-904 )
Date of Publication :
January 2015
Page Views :
284
Full Text Views :
1
|