No Detail Found!

Subjectivity lexicon construction for mining drug reviews.

Author(s)
Muhammad Zubair Asghar Institute of Computing and Information Technology, Faculty of Sciences, Gomal University, D.I.Khan, Pakistan
Aurangzeb Khan Institute of Engineering and Information Technology, Faculty of Sciences, University of Science and Technology, Bannu, Pakistan
Shakeel Ahmad Institute of Computing and Information Technology, Faculty of Sciences, Gomal University, D.I.Khan, Pakistan
Bashir Ahmad Institute of Computing and Information Technology, Faculty of Sciences, Gomal University, D.I.Khan, Pakistan
Abstract
Opinion mining or sentiment analysis aims at extracting the opinions or sentiments of online users about a service, product, and policy. In this paper, we propose a subjectivity lexicon for mining drug reviews available on different drug forums. The proposed method is based on the corpus of drug reviews. The lexicon takes the initial medical seed list as input, expands it with Senti Word Net synonyms and antonyms, attaching polarity score with each word, and finally filtering the final lexicon from unnecessary words. The comparative results show the efficiency of proposed method and it outperforms the existing approaches. The proposed approach achieves an accuracy of 74% on training corpus and 71% on testing corpus of drug reviews.
Publication Details
Page(s) 145-149
DOI DOI not available
Published Journal: Science International, Volume: 26, Issue: 1, Year: 2014

Request the full text from PASTIC

0

Citations

0

Downloads

36

Views