Investment Recommender System Model Based on the Potential Investors' Key Decision Factors

Area of Science:

• Artificial Intelligence
• Financial Technology
• Decision Support Systems

Background:

• Traditional investment recommendation systems often struggle with incomplete data and subjective investor decision factors.
• There is a need for intelligent systems that can process complex decision factors and provide personalized investment advice.

Purpose of the Study:

• To propose and evaluate a novel automatic recommender system for investment-type suggestions.
• To develop a model utilizing an adaptive neuro-fuzzy inference system (ANFIS) for enhanced investment decision support.

Main Methods:

• The proposed system employs an adaptive neuro-fuzzy inference system (ANFIS) integrating four key investor decision factors (KDFs).
• Data preprocessing utilizes the K-means technique, while ANFIS handles data evaluation and investment type recommendation.
• The system is designed to handle incomplete data and can incorporate expert feedback.

Main Results:

• The ANFIS-based system effectively predicts investor decisions based on their KDFs.
• Comparative analysis against existing investment recommender systems (IRSs) demonstrated the proposed system's accuracy and effectiveness.
• Root mean squared error (RMSE) was used to evaluate system performance.

Conclusions:

• The developed investment recommender system is reliable and effective for potential investors.
• The ANFIS approach provides a robust framework for personalized investment advice and decision-making.
• The system enhances the investment process by offering data-driven suggestions and supporting investor choices.