Modeling sentiment terminologies: Target based polarity phenomena

In recent few years, social networking site and information technology has become ubiquitous. This technology improvement is affecting the social and economic life of human beings more than anything. The social effects of the faster internet are directly linked to the dependence of today's generation on devices such as high speed internet mobile devices and social networking sites. Lately these social internet sites — such as Facebook, Amazon, and Twitter — have become a great aid for data mining and sentiment analysis. Various individuals and several companies are exploiting data from these sites to carry out their day to day business in an efficient manner. This efficiency can further be enhanced by determining sentiment on the basis of subject-centered targeted sentiment exploration with the help of uni-gram technique instead of utilizing non-subjective uni-gram or bi-gram methods. Experimental evaluation of our research work illustrates that the proposed method of subject based polarity phenomena of determining sentiment outperforms the previously proposed techniques.