ABSTRACT

Title is a compact representation of a document which distill the important information from the document. In this paper we studied the selection words as title words by using different learning approaches namely nearest neighbor approach (NN), Naive Bayes approach with limited-vocabulary (NBL), Naive Bayes approach with full vocabulary (NBF) and by using a term weighing approach (tf-idf). We compare the performance of these approaches by using F1 metric. We compare the F1 metric results both on English Script and Indic Script ' Telugu'. We concluded the influence of linguistic complexity in the process of Title word selection.