Hybrid of jellyfish and particle swarm optimization algorithm-based support vector machine for stock market trend prediction

Market prediction is a pivotal research domain within the financial market. The continuous evolution of information and communication technology has not only led to an exponential increase in data volume but has also introduced greater diversity in data formats. Thus, this study proposes a novel prediction model employing a hybrid of jellyfish and particle swarm optimization (HJPSO) algorithms. This hybrid model is designed to effectively manage the overwhelming volume of data, including technical indicators and financial news, while simultaneously optimizing the parameters of the support vector machine (SVM). In addition to its predictive capabilities, the study incorporates a rule extraction method, shedding light on the decision rules inherent in the SVM post-prediction. Computational results indicate that the proposed HJPSO-SVM is superior to existing algorithms in terms of accuracy and trading simulation. The incorporation of both stock indicators and news data emerges as a key factor contributing to enhanced predictive performance. This comprehensive approach reveals the significance of integrating diverse data sources for more robust market predictions.