Al-Powered Recommender Systems in Financial Market Analysis: A Narrative Review

The exponential growth of data has driven the use of artificial intelligence in financial analytics. Using massive amounts of data and complex algorithms, artificial intelligence (AI) systems can identify patterns, predict market movements, and make decisions with speed and accuracy that surpass human capabilities. This article provides a comprehensive review of the existing research on AI-based recommender systems in financial markets. This article aims to demonstrate the various machine learning (ML), deep learning (DL), and reinforcement learning (RL) techniques used in building and optimizing financial recommender systems and how these systems can make investment decisions smarter, more accurate, and more personalized. It also discusses the integration of various types of financial data (structured and unstructured) enabled by natural language processing (NLP), large language models (LLM), and knowledge graphs. The findings show that the integration of machine learning (ML), deep learning (DL), and reinforcement learning (RL) will enable financial institutions to provide more accurate, adaptive, and personalized investment recommendations. Although AI-based recommender systems have paved the way for a fundamental transformation in financial data analysis and investment decision-making, they also pose ethical, regulatory, and interpretability challenges. Finally, the study concludes that the next generation of AI-based recommender systems must not only be technically superior but also ethical and trustworthy. They must achieve a balance between analytical accuracy, transparency, and ethical responsibility to establish their role as decision-making tools in the future smart financial ecosystem.

کلیدواژه ها:

Artificial Intelligence (AI) ، Recommender Systems ، Financial Market Analysis ، Machine Learning (ML) ، Deep Learning (DL) ، Explainable AI (XAI)