IIIT Hyderabad Publications |
|||||||||
|
Towards Building a SentiWordNet for TamilAuthors: Abishek Kannan,Gaurav Mohanty,Radhika Mamidi Conference: 13th International Conference on Natural Language Processing (ICON-2016 2016) Date: 2016-12-17 Report no: IIIT/TR/2016/62 AbstractSentiment analysis is a discipline of Natural Language Processing which deals with analysing the subjectivity of the data. It is an important task with both commercial and academic functionality. Languages like English have several resources which assist in the task of sentiment analysis. SentiWordNet for English is one such important lexical resource that contains subjective polarity for each lexical item. With growing data in native vernacular, there is a need for language-specific SentiWordNet(s). In this paper, we discuss a generic approach followed for the development of a Tamil SentiWordNet using currently available resources in English. For Tamil SentiWordNet, a substantial agreement Fleiss Kappa score of 0.663 was obtained after verification from Tamil annotators. Such a resource would serve as a baseline for future improvements in the task of sentiment analysis specific to Tamil data. Full paper: pdf Centre for Language Technologies Research Centre |
||||||||
Copyright © 2009 - IIIT Hyderabad. All Rights Reserved. |