An Effective Feature Selection Based Classification model using Firefly with Levy and Multilayer Perceptron based Sentiment Analysis

Recently, Sentimental Analysis (SA) attains most important attention while making a decision when it used for extracting and classifying the sentiments existing in web-based reviews which are written by a customer. Here, SA is used as a sentiment classification (SC) issue where the reviews are classified as positive and negative factors which are based on internet reviews. This work projects an efficient SA technique for online reviews by the combination of Feature Selection (FS) as well as classification. The Firefly (FF) and Levy flights (FFL) models are used to extract features from web-based review and Multilayer Perceptron (MLP) is employed to classify the sentiments. For experimentation, a benchmark DVD dataset is employed to test the effectiveness of the presented model. The final outcome reveals that the proposed FF-MLP method attains effective classifying operation from every dataset on behalf of various performance metrics.