Domain Independent Joint Sentiment And Topic Detection UsingAFINN And Other Lexicons
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2015, Vol 17, Issue 4
Abstract
Abstract: Sentiment analysis and opinion mining field concentrates on automatically classifying sentiments of documents. But detecting mere sentiments does not provide one the sufficient knowledge concealed in the text.In this paper, we have evaluated joint sentiment and topic detection model (JST) to detect sentiment and topicsimultaneously from text. JST is evaluated for movie reviews and product reviews with the help of domainindependent prior information. AS no labeled data is required for training JST, it becomes highly portable toany domain. JST provides user more information regarding the text than mere sentiments.
Authors and Affiliations
Supriya Paul , Sachin deshmukh
Keywords
Related Articles
- EP ID EP106045
- DOI -
- Views 129
- Downloads 0
LTE Performance over Wimax in Term of Network Applications: An Implementation on E-Administration System
Today's application are based on the network as the main communication mediums for interconnecting between its peers. From this term, the importance of the network increased rapidly as the raise of technology and applica...
A Review onImage Mining Techniques and its application on asoftware BOND
Abstract: Image processing is one of the most researched areas in computer science and it finds numerousapplications in various fields like, medical research and diagnosis, geological research, crime investigation,and so...
Computerized Power Montoring System for EEDC Umuahia
Abstract: The computerized process in any facet of human existence or any sector of any economy promotesefficiency accuracy, effectiveness, improved service delivery and many more. This is also applies to thecomput...
Electricity, Computers, Innovation, Environment and Capacity Building: A Statistical Review
Abstract : The objective was to examine the frameworks needed to foster innovations and to mobilize innovation towards co - creating computers in the context of developing countries’ challenge of unreliable energ...
A Hierarchical and Grid Based Clustering Method for Distributed Systems (Hgd Cluster)
Abstract: In distributed peer-to-peer systems, huge amount of data are dispersed. Grouping of those data from multiple sources is a tedious task. By applying effective data mining techniques the clustering of distributed...