AI Text Extractor

This is a simple AI-powered pipeline that extracts structured information (like Names and Dates) from plain-text files using spaCy + Streamlit. The results are displayed in a web interface and also saved to a CSV file.

---

📂 Project Structure

Extractor/
│
├── scribe_portal.py    # Main Streamlit app
├── harvest_log.csv     # Output file (auto-created after run)

---

⚡ Requirements

• Python 3.9+
• pip (Python package manager)

Dependencies:

• streamlit
• spacy
• pandas
• spaCy English model: en_core_web_sm

---

🔧 Installation

1. Clone the repo

git clone https://github.com/your-username/AI-data-extractor.git
cd AI-data-extractor/Extractor

2. Create a virtual environment (recommended)

On Windows

python -m venv venv
venv\Scripts\activate

On Linux / macOS

python3 -m venv venv
source venv/bin/activate

3. Install dependencies

pip install -r requirements.txt
python -m spacy download en_core_web_sm

---

🚀 Running the App

Once inside the Extractor folder:

streamlit run scribe_portal.py

This will start a local web server. Open the link shown in terminal (usually http://localhost:8501).

---

📜 Usage

1. Upload the attached .txt file.

2. The app will extract Names and Dates.

3. Results are shown:

   • As JSON
   • As a table
   • Saved automatically in harvest_log.csv

---

📝 Example

Input file (sample_doc.txt):

Meeting Notes

On 15th September 2025, John Smith met with Alice Johnson in New Delhi to discuss the upcoming product launch. 
The deadline for the first phase is October 2, 2025. 
A follow-up meeting is scheduled for next Monday with Dr. Robert Brown. 

Prepared by: Sarah Williams

Extracted Output (CSV):

Names,Dates
John Smith,September 2025
Alice Johnson,"October 2, 2025"
Robert Brown,next Monday
Sarah Williams,

---