Ranking of Odia Text Document Relevant to User Query Using Vector Space Model

摘要

In this digital world, there is no lack of digitize text, image, audio and video for any subject but extracting the conceptual information relevant material is very difficult and achieved by different information retrieval(IR) technologies. Boolean Matching Model, Probabilistic Model, Extended Boolean Model, Fuzzy Set Model and Vector Space Model are some of the widely used models for information retrieval. Among these, vector space model (VSM) is one of the classical and widely used model which works based on the mathematical concepts of linear algebra. In VSM, queries and documents are the vectors in the highly multidimensional space and terms are used as dimensions to construct the index key to represent the documents. In this paper, we consider different approaches of vector space model for information retrieval and compare the results to find a neat understanding of term count model, classical vector space model and normalized vector space model. All the three models works based on the concept of term frequency (tf) and inverse document frequency (idf). Here the vector space retrieval technology we have applied on odia text to get an overview how VSM works on regional language like Odia. Odia (officially changed the name from Oriya to Odia in November, 2011) is one of the language mainly spoken by the people live in Odisha, the south east state of India. This Model can also be used for text summarization and data mining purpose.