This dataset is built for Nifty50 trade decision support and forecasting research by combining financial news, sentiment signals, and stock market data into one pipeline-ready format.
It supports end-to-end workflows such as:
News preprocessing and cleaning
News impact segregation (direct, sectoral, global)
Financial sentiment scoring (FinBERT)
Time-series forecasting with Temporal Fusion Transformer (TFT)
raw_dataset/Source-level scraped financial news files.… See the full description on the dataset page:
https://huggingface.co/datasets/mveen3/Six_Year_Indian_Stock_Market_Dataset-News_and_Ticker.
