AI-Powered Resume Screening & Candidate Ranking System

An intelligent, decision-support Machine Learning system built for HR teams and recruiters to automatically screen, score, and rank candidate resumes based on target job descriptions using Natural Language Processing (NLP).

🚀 Key Features Implemented

• Text Preprocessing Pipeline: Cleans unstructured resume text by removing URLs, emails, special characters, and non-informative stopwords.
• NLP Skill Extraction Layer: Utilizes spaCy to extract technical skills and industry keywords from unstructured profiles.
• Vectorization & Similarity Model: Transforms clean text documents into numerical vectors using TF-IDF Vectorization and calculates semantic alignment metrics using Cosine Similarity.
• Skill Gap Analysis: Computes structural skill differences mathematically, outputting explicitly what a candidate lacks relative to a job posting.
• Interactive UI Dashboard: Built an optional visual dashboard utilizing Streamlit to showcase rankings and performance charts cleanly for non-technical users.

🛠️ Tech Stack & Tools

• Language: Python
• Libraries: Scikit-learn, spaCy, NLTK, Pandas, NumPy
• Framework: Streamlit (UI Application)
• Environment: VS Code / Jupyter Notebook

📁 Repository Structure

├── data/                  # Repository datasets (Resume.csv / Monster Job Descriptions)
├── notebooks/             # Experimental sandbox
│   └── exploration_and_modeling.ipynb
├── src/                   # Production-ready Python core modules
│   ├── __init__.py        # Exposes module architecture packages
│   ├── preprocessor.py    # Text cleaning script routines
│   ├── extractor.py       # Entity and skill extraction layer
│   └── scorer.py          # Similarity vector computation engines
├── app.py                 # Interactive Streamlit Web UI application Dashboard
└── requirements.txt       # Software and environment package dependencies