🧠 Natural Language Processing Projects using Flask

This repository contains multiple complete NLP applications, all built using Flask with clean and responsive custom UIs.

🚀 Projects Included

✅ 1. Sentiment Analysis on Product Reviews

• Dataset: IMDb or Amazon Product Reviews
• Model: Logistic Regression, Naive Bayes
• Bonus: Visualization of positive/negative frequent words

✅ 2. News Category Classification

• Dataset: AG News Dataset
• Model: Logistic Regression, Random Forest, SVM, Neural Network
• Bonus: Word clouds, bar charts

✅ 3. Fake News Detection

• Dataset: Fake and Real News Dataset
• Model: Logistic Regression, SVM
• Bonus: Visualization with Word Cloud

✅ 4. Resume Screening using NLP

• Dataset: Resume + Job Descriptions Dataset
• Model: Semantic Search using Sentence Transformers
• Bonus: Matching percentage, named entity extraction, frontend resume upload

💡 Features

• Clean and fully functional frontend UIs for each app
• Complete backend logic using Flask
• Interactive model results with visualizations
• Upload resume and view job matches in real-time
• Modularized code for easy extension

---

🖥️ Folder Structure

nlp-projects/
│
├── task1\_sentiment\_analysis/
│   ├── app.py
│   ├── static/
│   ├── templates/
│   └── model.pkl
│
├── task2\_news\_classification/
│   ├── app.py
│   ├── static/
│   ├── templates/
│   └── model.pkl
│
├── task3\_fake\_news\_detection/
│   ├── app.py
│   ├── static/
│   ├── templates/
│   └── model.pkl
│
├── task4\_resume\_screening/
│   ├── app.py
│   ├── static/
│   ├── templates/
│   └── resume\_matcher.py
│
└── requirements.txt

⚙️ How to Run

1. Clone the repository
   git clone https://github.com/yourusername/nlp-projects.git

2. Navigate to a task directory
   cd task1_sentiment_analysis

3. Install dependencies
   pip install -r requirements.txt

4. Run the Flask server
   python app.py

5. Open in browser:
   http://localhost:5000

---

📈 What I Learned

• NLP techniques like tokenization, lemmatization, stopword removal
• Feature extraction using TF-IDF, CountVectorizer
• Multiclass and binary classification (Logistic Regression, Naive Bayes, SVM, Random Forest, Neural Network)
• Text similarity and semantic matching using Sentence Transformers
• Real-time resume screening and cosine similarity
• Designing beautiful Flask UIs with HTML/CSS
• Visualizing word frequencies using WordCloud and matplotlib
• Serving ML models via Flask APIs
• Full-stack development (frontend + backend) for NLP systems

---

Each task has a separate app.py—run the one you want to use.

---

🎯 Results & Performance

• All models were evaluated using accuracy and F1-score.
• Resume Screening used semantic similarity to deliver intuitive match scores.
• Visual feedback (word clouds, match highlighting) improves explainability.

---

💡 Future Enhancements

• Deploy to Heroku or Render
• Add login system to save past results
• Expand resume parsing with docx/pdf parsing libraries

---

📬 Contact

Created by Wahaj Asif Feel free to reach out or contribute via issues or pull requests.