Hybrid Multi-Instance Learning Model For Financial Market Prediction
DOI:
https://doi.org/10.64751/Keywords:
Financial Market Prediction, Hybrid Multi-Instance Learning, Machine Learning, Stock Market Forecasting, Financial Time Series Analysis, Predictive Analytics, Data Mining, Market Trend Analysis, Feature Extraction, Investment Decision SupportAbstract
Financial market prediction is a challenging task due to the dynamic, non-linear, and highly
volatile nature of stock price movements. Traditional statistical models often fail to capture
complex patterns present in financial data. To address these limitations, this project proposes
a Hybrid Multi-Instance Learning (MIL) based model for financial market prediction. The
proposed system integrates multiple data sources, including historical stock prices, technical
indicators, and financial news data, to improve prediction accuracy. Multi-Instance Learning
is employed to effectively handle grouped data samples, where each bag contains multiple
related instances representing different market signals. Additionally, NLP-based sentiment
analysis using VADER and SpaCy is incorporated to capture news-driven market signals.The
hybrid framework combines the strengths of multi-instance learning and XGBoost
classification to enhance prediction performance and robustness. Experimental results
demonstrate that the proposed model achieves improved accuracy compared to traditional
machine learning approaches. This system can assist investors and financial analysts in
making informed decisions by providing reliable market trend predictions
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
