DocumentCode
3764725
Title
Classification of drugs reviews using W-LRSVM model
Author
Asha S Manek; Kailash Pandey K;P Deepa Shenoy;M. Chandra Mohan; Venugopal K R
Author_Institution
Department of Computer Science and Engineering, Jawaharlal Nehru Technological University, Hyderabad, India
fYear
2015
Firstpage
1
Lastpage
6
Abstract
Opinion mining provided less opportunity to discuss their experiences about drugs so reviewing about it was difficult. Recent findings show that online reviews and blogs on drugs are important for patients, marketers and industries. Collecting the information for drugs from the website and analyzing is a challenge. A model is designed by proposing an algorithm which crawls information from the web to analyze reviews of drugs. Reviews were crawled for five different drugs using the algorithm. The W-Bayesian Logistic Regression and Support Vector Machine (W-LRSVM) model was trained for different split ratios to obtain the accuracy of 97.46%. Experimental results on reviews of five different drugs showed that the proposed model gave better results compared to other classifiers.
Keywords
"Drugs","Support vector machines","Logistics","Algorithm design and analysis","User interfaces","Classification algorithms","Blogs"
Publisher
ieee
Conference_Titel
India Conference (INDICON), 2015 Annual IEEE
Electronic_ISBN
2325-9418
Type
conf
DOI
10.1109/INDICON.2015.7443425
Filename
7443425
Link To Document