APSO BASED FEATURE EXTRACTION APPROACH AND SVM BASED CLASSIFICATION FOR SENTIMENTAL ANALYSIS

Abstract

The focal point of Sentimental Analysis is intolerance of estimation or else some sentiment so as to survive in the assessment. Based on the three categories, the execution of Sentiment Analysis has been done such as several supervised, un-supervised and machine learning algorithm. Nevertheless, each and every model has its own advantages as well as disadvantages. An effort has been given by our above mentioned proposed methodology to establish partial swarm intelligence based sentimental supervised methodology. Since huge quantity of data samples in sort to attain an appropriate characteristic data set, the above mentioned model has been used to achieve the highest finest characteristic set. Minimum Redundancy as well as Maximum Relevancy is used to calculate the evaluation of finest attribute set is achieved. By utilizing with the taxonomy of Support Vector Machine, the cataloging of the excerpt feature set has been proficient. The calculation of recall, precision, accuracy and f-measure are taken here for analysis.