Classifying User reviews of movie applications using Improved Logistic Regression.

In recent years review classification, analysis and prediction are one of the most common applications of sentiment analysis. It involves detection of sentiments on the reviews made by the users on social networking applications through opinion mining. In general, reviews can have positive, negative or neutral polarity indicators. For classification, the polarity indicators take the form of certain words and emotions that readily show the user’s sentiments. Existing works fall short of producing accurate classification results because of a two-class problem that affects the performance of evaluation parameters like precision, recall, accuracy and F-measure. Hence there is a need for an efficient classification technique which addresses two-class problems. This work proposes an Improved version of Logistic Regression ILR that is commonly used for sentiment analysis and classification. The proposed classification technique identifies and replaces the misspelled words in the sentence, supports count estimation and classification of reviews along with multiple independent words with similar meaning in parallel. The experimental results show the classification accuracy of the proposed technique to be more accurate compared to the existing logistic regression and naïve bayes classifiers.