,
Assistant Professor, Department of Computer and Information Science, Annamalai University , Chidambaram, Tamil Nadu , India
,
Assistant Professor, Department of Computer and Information Science, Annamalai University , Annamalai Nagar, Tamil Nadu , India
Department of Computer and Information Science, Annamalai University , Chidambaram, Tamil Nadu , India
Accurate prediction of stock market trends remains a challenging task due to high volatility, non-linearity, and the dynamic nature of financial time series data. Conventional statistical and machine learning typically do not provide consistent performance due to the fixed hyperparameter settings and the inability to adapt to a shifting market situation. In view of these, this paper will suggest a hybrid stock market trend prediction model that combines a Smooth Support Vector Machine (SSVM) and the Weighed Vulture Optimization Algorithm (WVOA) to optimize the hyperparameters and generalize better. Historical Nifty50 stock market data between January 2000 to April 2021 are used, and price-based attributes, trading volume, volatility, and engineered technical indicators are used. The WVOA algorithm dynamically optimizes critical SSVM parameters, which allows the exploration-exploitation strategy to be balanced and makes prediction more robust. Experimental results demonstrate that the proposed SSVM–WVOA model achieves an accuracy of 95.5%, precision of 94.2%, recall of 93.9%, F1-score of 94.1%, and ROC-AUC of 0.967, consistently outperforming conventional SVM, ARIMA, GRU, and LSTM models. The results verify that learning based on optimization is found to be a significant way to enhance the accuracy, stability and ability to generalize the forecasting. The suggested framework provides a computationally-efficient and scalable method to predict the trends of stock markets and can be successfully applied in the context of making informed decisions to invest in financial analytics systems and risk management.
This is an open access article distributed under the Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
0
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.