📄 Document Classifier (NLP | Deep Learning)

A deep learning–based document classification system built using TensorFlow, NLP preprocessing, and Streamlit.
This project classifies raw text and .txt files into predefined categories and also supports bulk file sorting into folders based on predicted labels.

---

🚀 Features

• ✅ Classify typed or pasted text
• ✅ Upload and classify .txt files
• ✅ Batch document classification from a folder
• ✅ Automatic file sorting into category folders
• ✅ Deep learning–based prediction using a trained neural network
• ✅ Interactive Streamlit web interface
• ✅ Displays class probabilities

---

🧠 Tech Stack

• Python 3.10
• TensorFlow / Keras
• NLTK
• Streamlit
• NumPy & Pandas
• Regex for text cleaning

---

📂 Project Structure

document-classifier-nlp/
│
├── Data/                               # Training data
├── New_Files/                          # New test files
├── UI_ScreenShot.png                   # App preview image
├── app.py                              # Streamlit app
├── basic_dl_doc_classification.ipynb   # Training notebook
├── meta_basic.json                     # Model metadata (categories, max_len)
├── news_basic_dl_model.h5              # Trained deep learning model
└── requirements.txt                    # Required dependencies

---

⚙️ Installation & Setup

1️⃣ Create Environment (Recommended)

conda create -n docclass python=3.10 -y
conda activate docclass

2️⃣ Install Dependencies

pip install -r requirements.txt

3️⃣ Download NLTK Resources (Auto on first run)

The app automatically downloads:

• stopwords
• punkt
• wordnet

---

▶️ Run the Streamlit App

python -m streamlit run app.py

Then open the browser link shown in terminal, for example:

http://localhost:8501

---

🗃️ Model & Tokenizer

This project uses:

• A trained deep learning model (.h5)
• A saved tokenizer (.pkl)

⚠️ These files are not included in the repository due to size and security. You must place your trained model and tokenizer in the project root to run predictions.

---

📊 Functional Modes

✅ Single Text Classification

• Type or paste text
• Upload .txt files
• Get predicted category + probability scores

✅ Bulk Folder Classification

• Input a folder path
• Automatically sorts .txt files into category folders

---

🖼️ Streamlit App Interface Preview

---

---