Text processing in information retrieval system using vector space model

Intelligent information retrieval is an important area in computer application in the 21st century. In Tamil documents, Morphology (separating noun and verb) concept is used to retrieve the text. In this paper we use new approach on text processing in the information retrieval system. So we can widen our search criteria namely, Vowel - Kuril (Short), Nedil (Long); Consonant - Vallinam (Hard), Mellinam (Soft) and Idaiyinam (Medium). So it would not wait for the entire word to enter; perhaps the searching process starts immediately after the first letter is entered, because the Database table is segregated into 5 components rather than a single Database table. So to minimise the time constraint, memory space and to do a smart search a new IR system is introduced. In the proposed system, searches can be divided into three categorise, namely (i) Main topic search (ii) Subtitle search and (iii) Keyword search. So the system would search quickly and retrieve required information only. In addition to that, every poem is displayed with related pictures. So users will show more interest and desire to read those poems. In the classical system, the user should give the exact word to retrieve the information. But in the proposed system the misspelled word could be corrected and the information can be retrieved, because internally the system has its own spell checker. This would be useful for Tamil literates, Tamil students, Tamil scholars, etc.