Analyze financial news headlines to infer market sentiment using Python, NLP, and machine learning (LSTM/LSTM+BERT).
This project scrapes and processes financial news headlines, applies NLP techniques to perform sentiment analysis, and explores how sentiment relates to market trends and investment decisions.
- Text Preprocessing: Cleans headlines by removing noise, tokenizing, and normalizing.
- Sentiment Classification: Implements models (e.g., RandomForest Classification) to label headlines as positive, negative, or neutral.
- Evaluation: Includes metrics like accuracy, ROC-AUC, confusion matrix, and possibly time-series correlation.
- Python 3.x
- Libraries:
pandas,numpy,scikit-learn,TensorFlow,Keras,transformers,TextBlob,Matplotlib,Seaborn
-
Clone the repository
git clone https://github.com/GeekyVishweshNeelesh/Python-Machine-Learning-Project.git cd Python-Machine-Learning-Project -
Create and activate virtual environment
python3 -m venv venv source venv/bin/activate # macOS/Linux venv\Scripts\activate # Windows
-
Install dependencies
pip install -r requirements.txt
- Open Project_Python_Financial_Market_News_Sentimental_Analysis.ipynb in Jupyter.
- Go through each section:
- Data loading & cleaning
- Exploratory Data Analysis (EDA)
- Preprocessing
- Training & evaluating models
- (Optional) Market sentiment correlation
- Execute all cells in order. Modify the data source or extend to live news feeds if needed.
Feel free to:
- Fork the repo
- Submit pull requests with improvements (e.g., model enhancements, code cleanup, dockerization)
- File bug reports or feature requests via GitHub Issues
Vishwesh Neelesh – Data Scientist
GitHub: GeekyVishweshNeelesh
Built with ❤️