Optimizing Resume Clustering in Recruitment - A Comprehensive Study on the Integration of Large Language Models (LLMs) with Advanced Clustering Algorithms

Aug 30, 2026

Upload by Maeve

This study investigates how Large Language Models (LLMs) combined with clustering algorithms can automate and optimize resume screening for recruitment. The research evaluates combinations of LLMs—BERT, RoBERTa, DistilBERT, and STSB RoBERTa—with K-means, DBSCAN, and hierarchical clustering, measuring grouping quality through content and semantic relevance. Findings show that LLM-clustering integration increases cluster precision and relevance, supporting faster, more accurate candidate evaluation. A web application is implemented to operationalize the best-performing combination for interactive recruiter workflows.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Introduction

Literature Reviews

Optimizing Resume Clustering in Recruitment: A Comprehensive Study on the Integration of Large Language Models (LLMs) with Advanced Clustering Algorithms Pobbathi Amaravathi, Safooraa Amjad Khan, Palle Sriharsha and Y. L. Malathi Latha Department of Information Technology, Stanley College of Engineering and Technology for Women(A) Hyderabad, Telangana, India [amaravathipobbathi@gmail.com](amaravathipobbathi@gmail.com), [safooraa.khan@gmail.com](safooraa.khan@gmail.com) [sriharshapalle33@gmail.com](sriharshapalle33@gmail.com), [drmalathi@stanley.edu.in](drmalathi@stanley.edu.in) Abstract—This study investigates the application of Large Language Models (LLMs) combined with clustering algorithms to automate and optimize the resume screening process in recruitment. The research evaluates the effectiveness of various LLMs such as BERT, RoBERTa, DistilBERT, and STSB RoBERTa in conjunction with clustering algorithms like Kmeans, DBSCAN, and hierarchical clustering. These combinations are assessed based on their ability to group similar resumes efficiently and accurately, considering factors such as content, context, and semantic relevance. Our research contributes to the field by rigorously analyzing the interplay between advanced NLP models and clustering techniques, identifying the optimal combinations for accurate and meaningful resume grouping. Additionally, we have developed a web application that integrates the most effective LLM-clustering combination, providing recruiters with an intuitive and interactive platform for analyzing clustered resumes. The results demonstrate that the integration of advanced NLP models with clustering techniques significantly improves the precision and relevance of resume clusters, leading to a more streamlined and efficient recruitment process. The final implementation shows promise in handling large datasets, enhancing the speed and accuracy of candidate evaluation and selection. Index Terms—Resume Clustering, Large Language Models, K-means, DBSCAN, Hierarchical Clustering, Recruitment Process, BERT, RoBERTa, Natural Language Processing. I. INTRODUCTION CLUSTERING is a fundamental principle in machine learning and data analysis, focused on grouping similar data points into clusters based on their characteristics. By identifying natural patterns and similarities within data, clustering helps reveal inherent structures and relationships that might not be immediately apparent. This technique is widely applied across various domains, including image recognition, customer segmentation, and natural language processing. In essence, clustering organizes data into groups, making it easier to analyse, interpret, and extract meaningful insights from large datasets. In the context of recruitment, clustering has immense potential to streamline and optimize the hiring process. Traditionally, HR professionals manually sift through hundreds or even thousands of resumes to identify the best candidates, a task that is not only labour-intensive but also subject to human error and bias. As companies grow and the volume of applications increases, this manual process becomes increasingly inefficient, often resulting in delays and missed opportunities to secure top talent. This is where clustering, combined with the power of artificial intelligence (AI) and natural language processing (NLP), can make a significant impact. Recent advancements in AI, particularly with Large Language Models (LLMs) like BERT, RoBERTa, and DistilBERT, have enabled machines to understand and generate human language with remarkable accuracy. These models excel at processing complex textual data, making them ideal for analyzing resumes and other job-related documents. When integrated with clustering algorithms, LLMs can automatically organize resumes into meaningful clusters based on factors like skills, experience, and qualifications. This not only accelerates the recruitment process but also enhances its objectivity by re

FAQs

What problem does the study address in recruitment?

The study targets the manual resume screening process, which is time-consuming, inefficient as application volumes rise, and prone to human error and bias.

Which LLMs and clustering algorithms are evaluated?

It evaluates LLMs such as BERT, RoBERTa, DistilBERT, and STSB RoBERTa alongside clustering methods including K-means, DBSCAN, and hierarchical clustering.

How is effectiveness of the LLM-clustering combinations assessed?

Effectiveness is assessed by how well combinations group similar resumes accurately and efficiently, considering factors like semantic relevance, computational efficiency, and scalability.

Aug 30, 2026

Upload by Maeve