Enhancing Stock Market Forecasting Through a Service-Driven Approach: Microservice System

Predicting stock market is a difficult task that involves not just a knowledge of financial measures but also the ability to assess market patterns, investor sentiment, and macroeconomic factors that can affect the movement of stock prices. Traditional stock recommendation systems are built as monolithic applications, with all components closely coupled within a single codebase. While these systems are functional, yet they are difficult integrating several services and aggregating data from diverse sources due to their lack of scalability and extensibility. A service-driven approach is needed to manage the growing complexity, diversity, and speed of financial data processing. However, microservice architecture has become a useful solution across multiple sectors, particularly in stock systems. In this paper, we design and build a stock market forecasting system based on the microservice architecture that uses advanced analytical approaches such as machine learning, sentiment analysis, and technical analysis to anticipate stock prices and guide informed investing choices. The results demonstrate that the proposed system successfully integrates multiple financial analysis services while maintaining scalability and adaptability due to its microservice architecture. The system successfully retrieved financial metrics and calculated key technical indicators like RSI and MACD. Sentiment analysis detected positive sentiment in Saudi Aramco's Q3 2021 report, and the LSTM model achieved strong prediction results with an MAE of 0.26 and an MSE of 0.18.

@article{Algarni2025,
  title     = {Enhancing Stock Market Forecasting Through a Service-Driven Approach: Microservice System},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {16},
  number    = {1},
  year      = {2025},
  publisher = {The Science and Information Organization},
  author    = {Asaad Algarni},
  doi       = {10.14569/IJACSA.2025.0160127},
  url       = {https://doi.org/10.14569/IJACSA.2025.0160127}
}