Overview:

The AI Resume Scorer is an Artificial Intelligence and Natural Language Processing (NLP) based web application that evaluates resumes automatically and provides a score, feedback, and improvement suggestions.

The system analyzes resumes based on various parameters such as skills, experience, education, keywords, grammar, and relevance to a specific job description.
It is designed to help students, recruiters, and job seekers improve their resumes and increase their chances of being shortlisted for interviews.

This project demonstrates real-world use of AI, NLP, and Machine Learning in recruitment and HR automation.

Objectives:

• To build an AI system that automatically evaluates and scores resumes.

• To help job seekers improve their resumes through AI-generated feedback.

• To assist recruiters in shortlisting candidates efficiently.

• To apply NLP and text analytics for document understanding and evaluation.

Key Features:

1. Resume Upload: Users can upload resumes in PDF or DOCX format.

2. AI-Powered Scoring: Assigns a score (out of 100) based on content quality, structure, and relevance.

3. Keyword Matching: Compares resume keywords with the job description (JD) to calculate match percentage.

4. Detailed Feedback: Provides AI-based suggestions to improve grammar, keywords, and formatting.

5. Skill Gap Analysis: Identifies missing skills required for the targeted job.

6. ATS Compatibility Check: Analyzes if the resume is Applicant Tracking System (ATS) friendly.

7. Multi-role Comparison: Allows users to check resume score against different job roles (e.g., Developer, Analyst).

8. Dashboard & Analytics: Displays overall score, keyword match rate, and readability level through charts.

9. User Authentication: Secure login for users and recruiters.

10. Admin Panel: Manage user submissions, track analytics, and view system logs.

Tech Stack:

• Frontend: HTML, CSS, Bootstrap, JavaScript

• Backend: Python (Flask/Django) / Node.js / PHP

• Machine Learning & NLP:

  • Libraries: NLTK, spaCy, scikit-learn, TextBlob, Transformers (BERT/GPT-based models)

  • Techniques: Keyword extraction, Named Entity Recognition (NER), Sentiment & grammar analysis

• Database: MySQL / MongoDB

• Additional Tools:

  • PyPDF2 / docx2txt for text extraction

  • TF-IDF or word embeddings for similarity measurement

  • Pre-trained resume datasets for model training