Requirements Engineering Process of Recommender System for the Nigerian Stock Market using Machine Learning Approach

Abstract

Stock prices are influenced by numerous complex and unpredictable factors, posing significant challenges for novice traders in identifying and utilizing pertinent information for optimal stock selection. This study proposes an innovative approach that integrates requirements engineering (RE) principles with machine learning (ML) techniques to predict stock prices in the Nigerian market based on historical data. The data, encompassing ten years of stock prices and volumes, as well as user preferences and budgets, was utilized to train and test an Artificial Neural Network (ANN) model. The model's predictions were then compared with those from Support Vector Regression (SVR), Linear Regression (LR), K-Nearest Neighbor Regression (KNNR), and Neural Network Regression (NNR). The implementation of the recommender system was carried out using the Python programming language, enabling the detection of patterns in stock price movements and providing stock suggestions based on these patterns to aid traders in making informed decisions. The NNR algorithm demonstrated superior performance, achieving an 85% confidence level. These results suggest that the proposed system can generate reliable predictions to guide stock trading decisions. This research offers a significant contribution to the field by demonstrating the effective synergy between RE and ML in enhancing stock trading strategies. The findings have important implications for both novice and experienced traders, providing a robust tool for navigating the complexities of stock market investments.