Stock prediction analysis by customers opinion in Twitter data using an optimized intelligent model

The earlier prediction of stock based on sentiment analysis is an essential factor in the finance-based intersection. However, the stock details prediction based on sentiment classification is not easy due to stock dataset variations; predicting stock details in an earlier stage requires good functions. Hence, the presented article has developed a novel Strawberry-based Bi-directional Recurrent Neural Model (SBRNM) to efficiently analyze the sentiments for stock prediction. The Twitter dataset is collected, and the preprocessing, feature selection, and sentiment analysis processes are done. The sentiments are classified with the help of strawberry fitness. Furthermore, the designed procedure is executed in the python environment with appropriate packages. Consequently, based on the sentiments, the stock details are predicted. The robustness of the projected method was verified by measuring the metrics like precision, F-measure, accuracy, execution time, error rate, and recall. Moreover, the presented model has attained an accuracy of 97.06%. In addition, to determine the proposed framework's effectiveness, SBRNM was compared with existing methods, and it has gained better results than other models.