Forecasting of the Stock Market Price Using LSTM-CNN Model with Various Representations of Collected Dataset

One of the most challenging tasks in computation is stock market forecasting. Many factors, such as physiological versus physical factors, rational versus irrational behavior, investor attitude, market rumours, etc., impact the projection. We examine the potential impact of data analysis on this industry. The productive market hypothesis expresses that when all market members and financial backers have prompt admittance to data about an organization and stock market developments, the impacts of those advancements have proactively been figured into the stock price. Thus, the claim is guaranteed to be the main spot value that can be utilized to predict a market's future behavior. This study applies Machine Learning (ML) calculations on authentic stock value information to appraise the future pattern, considering the previous stock cost as the completion of every contributing variable. ML methods can be used to make highly accurate forecasts and uncover patterns and insights we were unaware of. We give a structure that utilizes the LSTM-CNN model and the net development computation technique to investigate and project an organization's future development.