Research Work on Financial Technology
Selected Publications and Projects
A Systematic Review on Graph Neural Network-based Methods for Stock Market Forecasting
Abstract: Systematic review of graph-based approaches for stock market forecasting, covering classification, regression, and recommendation tasks. Reviews frameworks, features, datasets, models, and evaluation metrics. Analyzes results and highlights future research directions.
View Paper
An Approach Towards Stock Market Prediction and Portfolio Optimization in Indian Financial Sectors
Abstract: Proposes DR2TNet, a dynamic relation-aware network capturing intra- and intersector stock associations and temporal patterns. Utilizes a dynamically updated financial knowledge graph. Demonstrates superior returns over baseline models.
View Paper
A Hybrid Relational Approach Towards Stock Price Prediction and Profitability
Abstract: Introduces a hybrid relational model using peer market data and Random Forest Feature Permutation for stock price prediction in the US, India, and Korea. Combines temporal convolution and linear models, outperforming baselines in profitability.
View Paper
Deep Learning Techniques for Stock Market Forecasting: Recent Trends and Challenges
Abstract: Examines deep learning models for stock forecasting, including ANN, CNN, Seq2Seq, GANs, GNNs, and Transformers. Analyzes datasets, evaluation metrics, and outcomes. Highlights GNNs and Transformers for dynamic financial time series.
A Deep Learning Approach Towards Indian Stock Market Movement Prediction
Abstract: Proposes SM2PNet, integrating inter-firm correlations and temporal dependencies to predict NSE stock movements. Shows improved accuracy over other deep learning methods.