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

Abstract

In this research, we propose an automatic recommender system for providing investment-type suggestions offered to investors. This system is based on a new intelligent approach using an adaptive neuro-fuzzy inference system (ANFIS) that works with four potential investors' key decision factors (KDFs), which are system value, environmental awareness factors, the expectation of high return, and expectation of low return. The proposed system provides a new model for investment recommender systems (IRSs), which is based on the data of KDFs, and the data related to the type of investment. The solution of fuzzy neural inference and choosing the type of investment is used to provide advice and support the investor's decision. This system also works with incomplete data. It is also possible to apply expert opinions based on feedback provided by investors who use the system. The proposed system is a reliable system for providing suggestions for the type of investment. It can predict the investors' investment decisions based on their KDFs in the selection of different investment types. This system uses the K-means technique in JMP for preprocessing the data and ANFIS for evaluating the data. We also compare the proposed system with other existing IRSs and evaluate the system's accuracy and effectiveness using the root mean squared error method. Overall, the proposed system is an effective and reliable IRS that can be used by potential investors to make better investment decisions.