Journal of Digital Information Management K. Kiruthika, E.S. Samundeeswari 22 3 2024 https://10.6025/jdim/2024/22/3/83-90 https://www.dline.info/fpaper/jdim/v22i3/jdimv22i3_1.pdf Predicting stock price movements has been challenging yet crucial for investors and financial analysts. Fluctuations in stock prices are valuable economic indicators, providing insights into overall economic well-being, consumer confidence, and market sentiment. In this study, we evaluate the efficacy of three different machine and deep learning algorithms in anticipating stock price trends. We assess the performance of Logistic Regression, Random Forest Regression, and Long Short-Term Memory (LSTM) algorithms in forecasting whether a stock's price will rise or fall in the upcoming period, utilising historical stock price data as input features. Our findings demonstrate that while each algorithm exhibits varying degrees of predictive accuracy, LSTM networks stand out as they generally outperform Logistic Regression and Random Forest Regression in capturing the complex temporal dependencies inherent in stock price data. This suggests that LSTM networks, with their superior performance, hold significant promise as effective tools for stock price trend prediction, particularly in volatile and non-linear financial markets. This could be a game-changer in stock price prediction, instilling optimism about the future of stock market analysis.