An Optimized Bayesian Probabilistic Neural Network Model for Sentiment Analysis

An Optimized Bayesian Probabilistic Neural Network Model for Sentiment Analysis

Chaturbhujachari Chaturvedi1, Prof. Preetish Kshirsagar2

Abstract— Of late, big data and big data analytics has fund applications in diverse fields. Social media and allied applications is one such domain for research, where Artificial Intelligence has shown unprecedented impact. In this paper a mechanism has been proposed which can classify text data into classes of different sentiments. Data in the form of tweets has been used in this case. Pre-processing of raw data has been done prior to using it to train a neural network. A  Neural Network is then trained using the categories of the data which are tweets that correspond to happy, neutral and sad moods of the Twitter users. The Bayesian Regularization (BR) algorithm has been used for training the artificial neural network. It has been observed that this proposed technique achieves an MAE of 0.6 which is significantly lesser compared to previously existing work.

Keywords—Sentiment Analysis, Machine Learning, Deep Nets,Bayesian Regularization, Mean Square Error (MSE), Mean Absolute Error (MAE).