search for


Using genetic algorithm to build an investment optimization model based on fund clustering
Journal of the Korean Data & Information Science Society 2019;30:285-97
Published online March 31, 2019;
© 2019 Korean Data and Information Science Society.

Sung Seog Kang1 · Hyun Jun Lee2 · Kyong Joo Oh3

1Department of Investment Information Engineering, Yonsei University, 23Department of Industrial Engineering, Yonsei University
Correspondence to: Professor, Department of Industrial Engineering, Yonsei University, Seoul 03722, Korea. E-mail:
Received February 6, 2019; Revised February 25, 2019; Accepted February 27, 2019.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
The influence of funds in global financial markets has been increasing after the financial crisis on 2008 and low-interest rates. Funds setting amounts and asset values are also continuously on the rise in the domestic market, and a variety of funds with different strategies, characters, and main investment assets have been developed and released. In spite of these market trends, there are very few researches which provide a model to support investment on funds. the previous studies about investment are mostly related to stock markets, and the items traded in the security markets are very different from the general funds, so it is difficult to apply those research results to the fund market. This paper presents the new investment decision support model based on domestic fund data. The proposed model consists of the process of selecting the funds expected to achieve high performance and the process of optimizing the investment proportion for the selected funds. The model also utilizes the artificial intelligence methodology that has been effectively applied in recent financial data analyzes. As a result of the empirical analysis based on the data from July 2013 to June 2018, the presented model showed a higher rate of return than the risk-free interest rate and outperformed the classic equally-invested type of benchmark.
Keywords : Clustering, genetic algorithm, investment optimization model, k-means cluster analysis, self organizing map.