imtalha/AI-Powered-Resume-Screening-System

AI-Powered-Resume-Screening-System

Overview

This project is an AI-driven resume screening system that extracts key details from resumes, compares them with job descriptions, and computes a match score using NLP techniques. The system leverages Groq's Mixtral-8x7B-32768 model, Sentence Transformers, and Gradio for an interactive user interface.

Implementation Process

1. 1.Extracting Text from Resume (PDF)

The system uses pdfminer.six to extract text from uploaded PDF resumes.

• —The extracttextfrom_pdf() function reads the document and returns plain text.

1. 1.Extracting Structured Information from Resume

The extracted text is processed using Groq's LLM (Mixtral-8x7B-32768) to extract:

• —Name

• —Skills

• —Experience

• —Education

The extractresumedata() function sends the extracted text to the LLM and retrieves structured information.

1. 1.Matching Resume with Job Description

• —The system encodes both the resume details and job description into vector embeddings using SentenceTransformer('all-MiniLM-L6-v2').

• —Cosine similarity is computed between the two embeddings using pytorchcossim().

• —The matchresumewith_job() function returns a match score indicating candidate suitability.

1. 1.Building a User Interface with Gradio

The Gradio library is used to create a simple UI where users can:

• —Upload a resume (PDF format)

• —Input a job description

• —Get extracted resume information and a match score

The process_resume() function integrates all steps and returns results to the UI.

Technologies Used

• —Groq API: LLM for resume data extraction

• —pdfminer.six: Extract text from PDFs

• —Sentence Transformers: Convert text into vector embeddings

• —Torch: Compute similarity scores

• —Gradio: Build an interactive web interface

Installation & Setup

1. 1.Clone the Repository

git clone https://github.com/your-repo/AI-Resume-Screening.git
cd AI-Resume-Screening

1. 1.Install Dependencies

pip install -r requirements.txt

1. 1.Set Up Environment Variables

GROQ_API_KEY="your_api_key_here" 

1. 1.Run the Application

python app.py

Data Sources

• —Resumes: Provided by users in PDF format

• —Job Descriptions: Entered manually into the UI

Key Challenges & Solutions

1. 1.Extracting Accurate Resume Data

• —Challenge: Extracting structured information from raw text.

• —Solution: Used Groq LLM to parse text and extract fields accurately.

1. 1.Ensuring Resume & Job Matching is Effective

• —Challenge: Matching resumes with job descriptions using relevant skills.

• —Solution: Used Sentence Transformer embeddings for similarity computation.

1. 1.Handling Different Resume Formats

• —Challenge: PDF resumes have varied layouts.

• —Solution: Used pdfminer.six, which extracts text from diverse formats.

Future Improvements

• —Enhance extraction accuracy using fine-tuned LLM models.

• —Implement database storage for storing analyzed resumes.

• —Add multi-resume comparison to rank multiple candidates.