SVM-RideNN: hybrid support vector machine-rider neural network based sentiment analysis using product review

Nowadays, shopping on online sites has become most mainstream for users. Moreover, the social media comments and reviews serve a huge role in online shopping. The satisfactions of users have also been improved with the utilization of sentiment analysis (SA). In addition, the sentiment analysis determines the digital text, in which the messages are positive, negative, or neutral. SA has difficulty in examining sentiments that are expressed in different languages, dialects, or colloquialisms. Moreover, it faces trouble with slang, abbreviations, and emojis. To bridge this gap, a hybrid Support Vector Machine Rider Optimization Neural Network (SVM-RideNN)-based sentimental analysis is proposed in this research. This method aims to examine the type and identification of text representation, and performance of sentiment analysis. The input product reviews are obtained from the acquired dataset, and the essential features are refined. From the refined features, the sentimental analysis process is done using the SVM-RideNN. In addition, the metrics, namely precision, recall, F1-score, and rouge, are employed to validate the performance of the model, and it has attained the finest output of 91.8%, 94.9%, 93.3%, and 86.9%