Hybrid Framework of Machine Learning and Optimization for Adaptive Decision Making in Financial Markets

EasyChair Preprint 15679

Abstract

This paper introduces a novel hybrid framework combining machine learning (ML) and optimization techniques for adaptive decision-making in financial markets. The proposed framework leverages deep learning models, such as recurrent neural networks (RNNs), for accurate forecasting of financial data and employs particle swarm optimization (PSO) to determine optimal investment strategies. Experimental results on real-world financial datasets demonstrate the superiority of the hybrid approach in terms of prediction accuracy and profitability compared to traditional methods. The study highlights the potential of integrating ML and optimization to address the challenges of dynamic and uncertain financial environments.

Keyphrases: Optimization, Particle Swarm Optimization, Recurrent Neural Networks, financial markets, hybrid models, machine learning

@booklet{EasyChair:15679,
  author    = {Takumi Miyo and H Kung and Ryu Nao and Chi Zhang},
  title     = {Hybrid Framework of Machine Learning and Optimization for Adaptive Decision Making in Financial Markets},
  howpublished = {EasyChair Preprint 15679},
  year      = {EasyChair, 2025}}