CLUSTER-BASED RETRIEVAL USING LANGUAGE MODELING APPROACH: AN EXPLANATION

Abstract

Nowadays, one of the demands for computer system is capability to process text
md natural language automatically. Consequently, the development of algolithms that
€nabl€ computers to do such task has been one of the great challenges. Hence, any
mbstantial progress in this domain will have a strong impact on numeious applicationi
r.nging from information retrieval, information filtering, and intelligent agentiJo speech
rcmgnition, machine translation, and human-machine interaction [ro].
Information retrieval is a task to retrieve relevant documents in response to a
qurr-s b1'measuring similarity between documents in repositories and the query. In
ryaen-t !'ears, the meaning of the term 'similar' between documents and query has been
der oped. At first, a document is judged similar with the query merely bised on lexical
nnrtching sf the word between documents and query. Now, the term 'similar' is expanded
b 6e meening of the query. It means that the query is not necessarily expressed in the dment, to come up with thejudgment that a document is similar with the query.
Tn_ this paper I'm going to explain how it can be done, that the query is not
ttrPt"ssarib e>rpressed in the document, in text retrieval. But before it, I'll briefly describe re methods I used here. I'll close this paperby summarizing the explanation.