Sentiment Analysis-Based stock Market Prediction UsingOptimization Algorithm and Machine Learning

Stock prediction is the crucial model in online applications to sell the products to the customers based on their requirements. The main ‎problem in stock prediction is the high error rate and pattern recognition. In this paper, proposed a Flamingo Search Algorithm with ‎Random Forest (FSA-RF) for improving the performance of SMP with sentiment analysis. Moreover, historical stock dataset and stock ‎news dataset are collected from the net source and trained in the system. The datasets are updated to the developed FSA-RF model. Initially ‎data preprocessing is employed for synchronising the text through converting it into lowercase characters. Then polarity-based detection is ‎processed using stock news data for classifying the positive and negative polarities. Hereafter, sentiment analysis is executed to classify the ‎negative and positive sentiments based on the news dataset polarities. Sentiment analysis phase is used for accurate classification of positive, ‎neutral and negative sentiments that useful for predicting stock prices. Both datasets are combined and feature extraction is performed to ‎extract the relevant features related to stock market prices. Update the FSO fitness in random forest classification layer which classify the ‎stock prices accurately in the output layer. The designed model is implemented in python tool and the gained outcomes are validated with ‎other prevailing models in terms of accuracy, sensitivity, speciality, precision, and mean square error‎.