NASDAQ Stock Value Prognosis—A Comparative Scrutiny Using Deep Learning and Machine Learning Models

Predicting stock market trends is challenging due to the uncertainty and multiple factors that influence stock prices. This project uses deep learning and machine learning techniques to predict stock market trends. Based on stock market data from top IT companies listed on the NASDAQ stock exchange, the dataset is made up of 10 years of stock market data. There are five machine learning models considered in the study, including Decision Trees, Random Forests, Adaptive Boosters, Support Vector Classifiers, and Logistic Regression models, and deep learning methods included in the study such as Long Short-Term Memory. To gain insights, the first step is to conduct a data analysis of the past one-year data of the IT companies. Data for the last 60 days of the stock market were used only for testing and predicting, and data for the last 10 years of stock prices were used to evaluate the models. In order to train the LSTM model, the remaining nine years and ten months of data are used. LSTM models can be evaluated based on their accuracy by comparing the predicted values of the stock values over the last 60 days with the actual values of the stock during that period. The developed LSTM model demonstrates high accuracy when predicting stock values with minimal deviation. This project provides a valuable tool for stockholders and venture capitalists to make informed decisions by leveraging deep learning and machine learning techniques for stock market trend prediction.