High-Scale ATS and Semantic Resume Filtering using Django, NLP, and LLaMA3 - Groq Models

Abstract

An intelligent system that uses Natural Language Processing (NLP) and Machine Learning (ML) to automate resume classification is presented in this paper. Key resume features, such as education, skills, and job titles, are extracted and used to train models like Logistic Regression, SVM, Random Forest, and BERT. NLP preprocessing techniques, such as tokenization, stop word removal, and vectorization, prepare the text for analysis. The results of experiments show that these models improve classification accuracy and decrease the time needed for hiring. The system assists recruiters by ranking qualified candidates and eliminating applications that are not relevant, making the hiring process quicker and more efficient.

Introduction

The surge in job applications makes manual resume screening inefficient, inconsistent, and biased. To address this, intelligent automated systems using Artificial Neural Networks, Natural Language Processing (NLP), and Machine Learning (ML) are increasingly adopted for resume classification and candidate-job matching, achieving high accuracy (up to 94% in skill prediction).

Automated tools like Career Mapper analyze large datasets to provide objective resume feedback, aligning candidates’ profiles with industry expectations. Techniques such as K-Nearest Neighbors (KNN), Support Vector Machines (SVM), cosine similarity, and deep learning models (BERT, GPT-3) are used to rank and classify resumes based on relevance and compatibility with job roles.

The methodology involves preprocessing resumes (.pdf, .docx) through text extraction, normalization, named entity recognition, and skill extraction. Semantic matching using transformer-based embeddings (e.g., LLaMA3) enhances understanding by recognizing synonyms and context, surpassing simple keyword matching.

Machine learning algorithms applied include:

• Unsupervised learning (K-Means clustering): Groups similar resumes into job-relevant clusters without labeled data, aiding recruiters in visualizing candidate pools.

• Supervised learning (Random Forests, Decision Trees): Classifies resumes as suitable or not based on features like skill match, experience, and education, with transparent feature importance for recruiter insights.

The system is implemented using Python libraries (Scikit-learn, Pandas), Django for web interfaces, and incorporates CRUD operations to manage structured data. Overall, this AI-driven approach streamlines recruitment, reduces human bias, and improves fairness and efficiency in the hiring process.

Conclusion

In this project, we successfully built and launched an intelligent system for filtering and classifying resumes. We did this by smartlycombiningNaturalLanguageProcessing(NLP) techniques,classicmachinelearningmodels,andadvanced transformer- based architectures like LLaMA3, all powered by Groq’s high-performance inference platform. Oursystemtacklesseveralbigchallengesintoday\'shiringprocesshead-on: 1) HandlingMessy ResumeData:Itskillfully pullsoutinformationfromallsortsofdiverseandunstructured resume formats. 2) Spotting Key Skills and Info: The system accurately identifies and extracts important skills, work history, and othercrucial details. 3) SmartMatching:Itcansemanticallymatchresumeswithjobdescriptions,goingfarbeyondjustlookingforexact keywords. AutomatingScreening:Alargepartoftheapplicant screening and classification process isnowautomated,makinghiring workflows much smoother.

References

[1] Prof. DikshendraSarpate,PrarthanaKolhe,SrushtiKalbhor,SanchiYehalegaonkar (2024)“AIEnhancedSkillMatcher” [2] ArtificialIntelligent&DataScience,ZealCollegeOfEngineeringandResearch,India,pp.2582-52082024. [3] Ms. Y. Sowjanya, MareddyKeerthana, Pulluri Suneeksha, Dorgipati Sai Sri Harsha (2023) “Smart Resume Analyzer” Assistant Professor at Department of IT, Anurag Group of Institutions, Hyderabad 2Department of IT, Anurag Group of Institutions, Hyderabad, pp. 409-418 2023. [4] George Stalidis and Selini Kyriazidou (2023) “Job Role Description and Skill Matching in a Rapidly Changing Labor Market Using Knowledge Engineering” A. Kavoura et al. (eds.), Strategic Innovative Marketing and Tourism, Springer Proceedings in Business and Economics, Thessaloniki, Greece,2023. [5] LavKumar,KarthikPenikalapati,SudheerKumarReddyGowrigari(2023)“ResumeMatchingFrameworkViaRanking [6] AndSortingUsingNLPAndDeepLearning”SalesforceInc,India,pp.2582-52082023. [7] Riya Pal, Shahrukh Shaikh, Swaraj Satpute and Sumedha Bhagwat (2022) “Resume Classification using various Machine Learning Algorithms” Ramrao Adik Institute of Technology and D.Y. Patil Deemed to be University, Ramrao Adik Institute of Technology, Nerul, Navi Mumbai, India, ITM Web of Conferences 44, 03011,2022. [8] I.Umoren,N.Akwang,S.Inyang,A.Afolorunso,andG.James,\"NLP-semanticmachinelearning-basedsystem for intelligent classification ofprofessional skill-setsfor efficienthumanresourcemanagementprocess,\"Iota, vol. 5,no. 1, 2025. [Online]. Available:https://doi.org/10.31763/iota.v5i1.882 [9] V.Lai,K.J.Shim,R.J.Oentaryo,P.K.Prasetyo,C.Vu,E.-P.Lim,andD.Lo,\"CareerMapper:AnAutomated Resume Evaluation Tool,\" in Proc. IEEE Int. Conf. Big Data (Bigdata), 2016. [Online]. Available: http://research.larc.smu.edu.sg/careermapper [10] R.T.Fareed,V.Rajath,andS.Kaganurmath,\"Resume classificationandrankingusingKNNandcosinesimilarity,\"Int. [11] J.Eng.Res.Technol.(IJERT),vol.10,no.8,pp.192–195,Aug.2021.[Online].Available:http://www.ijert.org [12] B. Kinge, S. Mandhare, P. Chavan, and S. M. Chaware, \"Resume Screening using Machine Learning and NLP: A Proposed System,\"Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol., vol. 8, no. 2, pp. 253–258, Mar.-Apr. 2022. doi: https://doi.org/10.32628/CSEIT228240

Copyright

Copyright © 2025 Nita Chaudhary, Priyanka Sharma. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.