MACHINE LEARNING CLUSTERING IN FINANCIAL MARKETS: A LITERATURE REVIEW

Abstract

This paper aims to present a concentrated overview of innovative research in machine learning clustering techniques as applied to to different facets of financial markets and stock market investing. Research on techniques such as K-Means Clustering or Agglomerative Hierarchical Clustering and their derivatives play a pivotal role in augmenting the stock market research and investment strategies. Wheter it is time series clustering and prediction, portfolio selection and optimization, or risk management, machine learning clustering has potential to enhance already existing processes by improving performance, reducing time spent on repetitive tasks or mitigating human errors. A truly innovative tool in the investor’s toolset, it is imperative to not overlook its limitations, such as the necessity of selecting the appropriate technique for specific datasets, or the need for human supervision to maximize its utility and insights extracted.