Stock price trend prediction plays a vital role in financial decision-making, helping investors and traders anticipate market movements. Due to the complexity and volatility of stock markets, traditional statistical models often struggle to deliver accurate forecasts. This project explores the use of machine learning techniques to predict stock price trends based on historical data and technical indicators. By applying algorithms such as Support Vector Machines (SVM), Random Forest, and Long Short-Term Memory (LSTM) networks, the system aims to classify future trends as upward, downward, or neutral. The dataset includes features like open, high, low, close (OHLC) prices and trading volume, which are preprocessed and used for model training and testing. The expected result is a system that predicts movements in stock prices over the short term so that investors can make decisions based on accurate information. The study also focuses on measuring different accuracy indicators along with evaluating the model’s performance in terms of precision, recall, and F1-score. It is anticipated that the results will enrich the domain of financial forecasting while also serving as an initial step towards more elaborate research to build ensemble models.

This project demonstrates the effectiveness of machine learning techniques—particularly Decision Tree, Support Vector Machine (SVM), Random Forest, and Long Short-Term Memory (LSTM) networks—in predicting stock price trends based on historical and technical data. The study highlights how data preprocessing, feature extraction, and parameter tuning significantly enhance model accuracy and reliability. The Decision Tree algorithm provided a clear, interpretable baseline for trend classification, while Random Forest improved performance by reducing overfitting through ensemble learning. SVM showed strong results in handling high-dimensional data, and LSTM outperformed other models in capturing sequential and temporal dependencies within stock data. The findings confirm that integrating these models enables better understanding of market dynamics and more precise prediction of upward, downward, or neutral price movements. This system can serve as a valuable decision-support tool for investors, helping them make informed and timely financial choices. Overall, the project contributes to the advancement of AI-driven financial forecasting and provides a foundation for future research in hybrid and real-time predictive analytics.

[1]Noviaristani, Siska (2021) Contemporary Research on Business and Management. Available via Taylor & Francis Routledge and academic libraries (ISBN 978-1-032-05097-3). [2]Dey, S., & Das, A. (2020) Forecasting stock market using hybrid machine learning technique, Journal of Physics: Conference Series, 1706(1), 012012 Accessible via IOP Science. [3]Molnar, Christoph (2020) Interpretable Machine Learning.Freely available online via the author’s website (searchable via Google Scholar or ResearchGate). [4]Chen, T., & Guestrin, C. (2016) XGBoost: A scalable tree boosting system, KDD ’16 Official KDD PDF: ([turn0search1]) arXiv preprint: ([turn0search3]) DOI 10.1145/2939672.2939785 ([turn0search1], [turn0search5]) [5]Nguyen, T., Shirai, K., & Velcin, J. (2015) Sentiment analysis on social media for stock movement prediction, Expert Systems with Applications, 42(24), 9603–9611 Available via Elsevier ScienceDirect. [6]Patel, J., Shah, S., Thakkar, P., & Kotecha, K. (2015) Predicting stock and stock price index movement using Trend Deterministic Data Preparation and machine learning techniques. PDF (Nirma University): Par.CSE Nsysu ([turn0search4]). ResearchGate preview & DOI 10.1016/j.eswa.2014.07.040 ([turn0search0], [turn0search14]) ScienceDirect abstract: ([turn0search2]) [7]Bontempi, G., Taieb, S. B., & Le Borgne, Y. A. (2013) Machine learning strategies for time series forecasting, EBISS, pp. 62–77 Accessible via SpringerLink. [8]Atsalakis, G. S., & Valavanis, K. P. (2009) Surveying stock market forecasting techniques – Part II: Soft computing methods, Expert Systems with Applications, 36(3), 5932–5941 Available via Elsevier ScienceDirect. [9]Kumar, M. R., & Ravi, V. (2007) Bankruptcy prediction in banks and firms via statistical and intelligent techniques – A review, European Journal of Operational Research, 180(1), 1–28 Available through Elsevier ScienceDirect. [10]Huang, W., Nakamori, Y., & Wang, S. Y. (2005) Forecasting stock market movement direction with support vector machine, Computers & Operations Research, 32(10), 2513–2522 Accessible through Elsevier ScienceDirect. [11]Zhang, Y., & Zhou, Z.-H. (2004) Multiclass learning problems: A decision tree approach, Pattern Recognition, 37(4), 721–728 Access via Elsevier ScienceDirect (search title + DOI). [12]Suen, C. Y., & Lam, L. (1990) Multiple decision trees for pattern recognition, IEEE Transactions on Computers, 39(4), 464–475 Available via IEEE Xplore or academic PDF repositories.