The Stock Market and its tendencies are very erratic in the finance sector. Recent studies have demonstrated the enormous influence that information stories and social analyses may have on investors’ perceptions of the financial marketplace. Consequently, The goal of this study is to investigate how stock market news mood movements are related use of data from various news outlets, business organization publications, and commercial enterprise websites. By utilizing prior information about the model’s structure, In comparison to the auto-regressive integrated moving average model as well as vector autoregression technique, this will be study offers an implementation of The Bayesian Structural Time Series (BST) Model that has greater transparency and permits better handling of uncertainty. The assumption of linearity is one of the model’s major flaws. A nonlinear model, The Long Short Term Memory model had been potential to represent a variety of nonlinear properties present in the data set. The suggested approach uses a hybrid model that incorporates the long-short-term memory and Bayesian Structural Time Series models, as well as a regression component that gathers data from different news outlets to find market predictors. The suggested approach recognizes unexpected behavior or abnormal patterns in stock price movement, making it preferable to previous methods.