Search

The study reviews the application of machine learning tools in financial investment portfolio management, focusing on cluster analysis for asset allocation, diversification, and risk optimization. The paper aims to explore the use of clustering analysis to broaden the concept of portfolio diversification beyond traditional volatility metrics. An open dataset from Yahoo Finance includes a ten-year historical period (2014–2024) of 130 actively traded securities from international stock markets used. Dataset selection prioritizes top liquidity and trading activity. Python analytical tools were employed to clean, process, and analyze the data. The methodology combines classical Markowitz optimization with clustering analysis techniques, highlighting variance-return trade-offs. Various asset characteristics, including annualized return, standard deviation, Sharpe ratio, correlation with indices, skewness, and kurtosis, were incorporated into the clustering models to reveal hidden patterns and groupings among financial assets. Results show that while clustering enhances insights into asset diversity, classical approaches remain historically superior in optimizing risk-adjusted returns. This study concludes that clustering complements, rather than replaces, classical methods by broadening the understanding of diversification and addressing many diversity factors, such as metrics of the technical, graphical, and fundamental analysis. The paper also introduces the diversity rate based on clustering, which measures the variance balance by all features within and between clusters, providing a broader perspective on diversification beyond traditional metrics. Future research should investigate dynamic clustering techniques, integrate fundamental economic indicators, and develop adaptive models for effective portfolio management in evolving financial markets.

Microbiome research, the study of microbial communities in diverse environments, has seen significant advances due to the integration of deep learning (DL) methods. These computational techniques have become essential for addressing the inherent complexity and high-dimensionality of microbiome data, which consist of different types of omics datasets. Deep learning algorithms have shown remarkable capabilities in pattern recognition, feature extraction, and predictive modeling, enabling researchers to uncover hidden relationships within microbial ecosystems. By automating the detection of functional genes, microbial interactions, and host-microbiome dynamics, DL methods offer unprecedented precision in understanding microbiome composition and its impact on health, disease, and the environment. However, despite their potential, deep learning approaches face significant challenges in microbiome research. Additionally, the biological variability in microbiome datasets requires tailored approaches to ensure robust and generalizable outcomes. As microbiome research continues to generate vast and complex datasets, addressing these challenges will be crucial for advancing microbiological insights and translating them into practical applications with DL. This review provides an overview of different deep learning models in microbiome research, discussing their strengths, practical uses, and implications for future studies. We examine how these models are being applied to solve key problems and highlight potential pathways to overcome current limitations, emphasizing the transformative impact DL could have on the field moving forward.