Due to the variable and non-linear structure of financial stock markets, correctly forecasting stock prices is a stressful job. For the investment in stock market, which is an extremely composite & difficult process, prediction plays a very essential role. In fact, the latest extension in financial stock market forecast technology is machine learning, which generates the forecast through training on the price of present financial stock market indicators and their history. Machine learning uses different models in order to come up with correct predictions. In this extensive research work, all the models employed for evaluating the future movement in the banking industries are accurately computed. Two of these models comprise Back Propagation Neural Network (BPNN) & Support Vector Machine (SVM). When comparing with SVM and BPNN, the later usually gains higher accuracy than that of the former in each of the evaluated criteria. In terms of Mean Absolute Error (MAE), for example, BPNN model demonstrates much lower values across all the banks that were investigated, when compared to SVM. MSE (Mean Squared Error) values for the BPNN are significantly less for several data sets and the results prove higher calibration of the forecasts. Moreover, on the analysis of the R-squared (R2) values, there is evidence for a more accurate estimate of fluctuations in the stock market movement by the BPNN. Evidently, such a view provides the stakeholders with a convincing tool that aids them more safely and economically navigate the world of issues related to the banking sector investments.