---

Financial Text Intelligence

Financial Text Intelligence is an AI-powered system that analyzes financial news text and predicts the most relevant financial category using Natural Language Processing (NLP) and Machine Learning.

The project includes a complete machine learning pipeline and an interactive web application built with Streamlit.

---

Features

• Financial news text classification
• NLP text preprocessing
• TF-IDF feature extraction
• Logistic Regression machine learning model
• Category probability visualization
• Important keyword explanation
• Interactive web interface using Streamlit

---

Technologies Used

Programming Language

• Python

Libraries

• pandas
• scikit-learn
• nltk
• matplotlib
• Streamlit

Machine Learning

• TF-IDF Vectorization
• Logistic Regression Classifier

---

Project Structure

financial-text-intelligence
│
├── data
│   └── FinSen_US_Categorized.csv
│
├── results
│
├── src
│   ├── main.py
│   ├── model.py
│   ├── preprocessing.py
│   └── visualization.py
│
├── app.py
├── requirements.txt
└── README.md

---

How to Run the Project

Install dependencies

pip install -r requirements.txt

Train the model

cd src
python main.py

This step trains the machine learning model and saves the trained model files.

---

Run the web application

streamlit run app.py

Then open the application in your browser:

http://localhost:8501

---

Example

Input

The Federal Reserve signaled that interest rates may remain high due to inflation.

Output

Predicted Category: Credit and Lending
Confidence: 0.67

The system also displays:

• category probability distribution
• top predicted categories
• important keywords influencing the prediction

---

Future Improvements

Possible improvements include:

• transformer-based models (BERT)
• improved dataset balancing
• financial risk detection
• deploying the application online

---

Author

GitHub: https://github.com/SIRILEKKALA

---