Sentiment Analysis of Sinhala News Comments.

Abstract

Opinion mining or sentiment analysis extracts the users’ opinions, sentiments and demands from the subjective texts in a specific domain and distinguishes their polarity. Such subjective text includes product and service reviews, twitter messages, and comments on online news articles Although there is ample sentiment analysis research for languages such as English and Chinese, minimal research can be found for Sinhala. One reason for this is lack of datasets annotated with sentiment information. However, with the increase of online news sites, there are ample news comments available. These resources can be utilized to create a human-annotate dataset. Analyzing such news comments is beneficial to many parties including the general public, government and intelligent services. The money of this project was used to create following sentiment analysis datasets, and other supporting Sinhala datasets. Out of these, the first already resulted in a publication (see attached). Publications of the other two are still being prepared, 1. Sinhala code-mixed text classification dataset 2. Sinhala Named Entity recognition dataset 3. Sinhala co-reference dataset