Hybrid Quantum Fuzzy Neural Network Approach- Based SNS Sentimental Analysis for Stock Market Prediction

The growing reliance on artificial intelligence (AI) and big data in financial market analysis demands innovative methodologies to improve the accuracy of market trend predictions. Traditional models, which primarily focus on numerical stock market indicators, often fail to account for the psychological and sentiment-driven factors that significantly influence market behavior. This paper presents a hybrid approach that integrates sentiment data from social media platforms, such as Twitter, with conventional stock market indices using quantum fuzzy neural networks (QFNNs). By harnessing the computational power of quantum processors, the adaptability of fuzzy logic, and the pattern recognition capabilities of neural networks, the proposed system achieves enhanced predictive accuracy and provides deeper insights into market dynamics. The QFNN model demonstrates remarkable performance, with classification models like the support vector classifier (SVC) using radial basis function kernels achieving an accuracy of 98% and an F1-score of 97%. Additionally, the random forest (RF) model attains even higher accuracy at 99%, paired with an F1-score of 99%. The robustness of the model is further validated through receiver operating characteristic curves, with area under the curve scores reaching 1.0 for both SVC and RF models, underscoring their exceptional discriminatory power. This integration of qualitative sentiment analysis with quantitative market data represents a significant paradigm shift in financial forecasting, addressing many limitations of classical methods. Beyond stock market prediction, the study highlights the broader applicability of QFNNs in domains requiring large-scale data analysis and decisionmaking under uncertainty. The findings underscore the transformative potential of quantum computing and fuzzy logic in advancing AI-driven economic modeling and shaping the future of financial analytics.