AI-Powered Resume Shortlisting Agent (Skill-Aware): Resume Ranking using TF-IDF, OCR and Skill Coverage

This paper presents an AI-powered resume shortlisting prototype designed to assist recruiters in the initial screening of job applicants. The application is developed using Python and Streamlit. It extracts text from PDF resumes using PyMuPDF and applies Tesseract OCR to scanned documents. A rule-based validation mechanism identifies likely resume documents. Skills are extracted using a predefined vocabulary of 47 terms. The system calculates TF-IDF cosine similarity between the job description and each resume. It also calculates skill coverage based on the required skills identified in the job description. These two measures are combined using a weighted scoring approach, with 60% weight assigned to text similarity and 40% to skill coverage. The application ranks candidates and displays matched skills, missing skills, scores, and review recommendations. User authentication is implemented using SQLite and PBKDF2 password hashing. Functional testing was conducted using synthetic resume documents. The ranking accuracy has not been independently evaluated against human recruiter rankings. The system is intended as a decision-support tool, and all hiring decisions must be made by humans.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-16
DOI
https://doi.org/10.5281/zenodo.24277310
Primary Topic
AI and HR Technologies
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

AI-Powered Resume Shortlisting Agent (Skill-Aware): Resume Ranking using TF-IDF, OCR and Skill Coverage

Aniruddha Gaikwad, Sairaj Salunke, Sudharshan Neel
4 links
Zenodo (CERN European Organization for Nuclear Research)
AI and HR Technologies
preprint

AI-Powered Resume Shortlisting Agent (Skill-Aware): Resume Ranking using TF-IDF, OCR and Skill Coverage

Aniruddha Gaikwad, Sairaj Salunke, Sudharshan Neel
preprint en

Abstract

This paper presents an AI-powered resume shortlisting prototype designed to assist recruiters in the initial screening of job applicants. The application is developed using Python and Streamlit. It extracts text from PDF resumes using PyMuPDF and applies Tesseract OCR to scanned documents. A rule-based validation mechanism identifies likely resume documents. Skills are extracted using a predefined vocabulary of 47 terms. The system calculates TF-IDF cosine similarity between the job description and each resume. It also calculates skill coverage based on the required skills identified in the job description. These two measures are combined using a weighted scoring approach, with 60% weight assigned to text similarity and 40% to skill coverage. The application ranks candidates and displays matched skills, missing skills, scores, and review recommendations. User authentication is implemented using SQLite and PBKDF2 password hashing. Functional testing was conducted using synthetic resume documents. The ranking accuracy has not been independently evaluated against human recruiter rankings. The system is intended as a decision-support tool, and all hiring decisions must be made by humans.

Zenodo (CERN European Organization for Nuclear Research)
MIT Art, Design and Technology University (IN)
AI and HR Technologies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon