Data Analysis and Machine Learning algorithms play very vital role in various fields and application including forecasting and investment in stock markets. Forecasting stock market patterns is very useful because correctly forecasting stock prices may contribute to moneymaking returns by making the right decisions. Several analysts and researchers have always been mindful of stock market predictions. In the current domain of framework, various techniques are developed to predict stock market trends. But, due to non-stationary, blasting, and volatile results, stock market forecasting is a major challenge, and thus making it quite difficult to take decisions regarding investment for spinning the big money, and hence stock price analysis and prediction is a promising and exciting challenge. Academic researchers have upgraded many predictive models to predict stock prices. Nonetheless, there are several negative aspects in the previous methods after analyzing the past research, namely, strict quantitative assumptions are essential, human interactions are essential in the forecasting process and an acceptable scope is difficult to be identified. Due to the problems identified, proposing a hybrid and integrated method to predict the stock market prices seems to be promising at present as this approach can blend many different approaches to improve the overall performance of the model. After studying many research papers over the last decade, it may seem that hybrid method framework looks more promising to provide more desirable results as compared to the individual methods.