Skip to content

puruz14/Unifyr

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Unifyr
Unifyr

AI-driven recruitment platform streamlining expert-candidate matching.

Project Title

UNIFYR - AI-Powered Expert Panel Matching System
Problem Statement ID: 1654
Theme: Smart Automation
Category: Software
Team Name: Null Pointers

Introduction

In the competitive job market, making sure candidates are matched with the right experts for interviews is crucial. UNIFYR aims to automate the expert matching process using AI, making recruitment faster, fairer, and more efficient. As a team of passionate college students, we are developing a solution that utilizes Natural Language Processing (NLP) and similarity algorithms to achieve this goal.

Objectives

  1. Automate the Matching Process: Leverage AI to automatically match candidates with the most suitable experts based on their skills.
  2. Make Recruitment Faster and Easier: Reduce the time and effort needed to assemble the right expert panel for candidate evaluation.
  3. Ensure Fair Matches: Utilize algorithms to ensure that matches are skill-based, thereby reducing biases in the recruitment process.

Technical Approach

Our approach involves using multiple AI techniques to extract key information from candidates' CVs and match them with experts efficiently:

  • Natural Language Processing (NLP): Used to extract relevant information like skills, experience, and qualifications from candidate profiles.
  • K-Means Clustering: Groups candidates and experts based on the similarity of their skills, helping to form potential matches.
  • Cosine Similarity with LDA (Latent Dirichlet Allocation): Measures the similarity between candidate skills and expert profiles, ensuring closely related matches.
  • FAISS Search (Facebook AI Similarity Search): Accelerates the process by quickly finding the closest matches within large datasets.
  • Bipartite Graph Matching: Pairs candidates and experts in a way that maximizes match quality across the entire set.
  • Hybrid Recommendation Algorithm: Combines all the above methods to calculate a similarity score for each match, optimizing the pairing process.

Feasibility

The project is technically feasible using well-established tools and frameworks like Scikit-learn, Gensim, and Faiss. The system is designed to scale and handle a large volume of data.

Impact and Benefits

UNIFYR will transform the recruitment process by:

  • Streamlining Expert-Candidate Matching: Reduces manual effort, enhances accuracy, and accelerates the matching process.
  • Reducing Bias: Ensures fair and impartial evaluations using AI-based algorithms.
  • Optimizing Resources: Saves time and money by automating the recruitment workflow.
  • Enhancing Decision-Making: Aligns candidate skills with organizational needs, promoting diversity and effectiveness in candidate selection.
  • Modernizing Government HR Systems: Utilizes advanced AI technologies to bring efficiency and innovation to recruitment processes.

🔗 Links

🤖 Tech-Stack

Front-end

  • React
  • Vite
  • Tailwind CSS

Back-end

  • NodeJS
  • ExpressJS
  • Flask

Database

  • MongoDB

Machine Learning

  • Scikit-learn
  • Gensim
  • FAISS

💸 Applications

Unifyr provides several key benefits:

  • Streamlined Expert-Candidate Matching: Automates the matching of candidates with subject matter experts, reducing manual workload and improving match accuracy.
  • Bias Reduction: Uses algorithms to minimize human biases, ensuring fair and impartial evaluations.
  • Resource Optimization: Reduces time and costs associated with manual recruitment tasks, freeing up HR resources for strategic activities.
  • Enhanced Decision-Making: Aligns candidates' expertise with organizational needs, ensuring the best fit for specific roles and missions.

🛠 Project Setup & Usage

To set up and run the project locally:

Clone the Git Repository

git clone https://github.com/VaibhavWalujkar04/Unifyr.git
cd Unifyr

Start Front-end

cd client
npm install
npm run dev

Start Back-end

cd server
npm install
npm run dev

Set up Python Environment

cd utils
pip install -r requirements.txt
python app.py

👨‍💻Team Members

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • JavaScript 55.6%
  • Jupyter Notebook 33.7%
  • Python 10.4%
  • Other 0.3%