Intelligent Resume Parsing and Job Recommendation via Web-Based CV Analysis System

Abstract

The process of matching job applicants to suitable job openings is typically inefficient due to unorganized resume formats and absence of customized job search facilities. This paper presents a light-weight web-based system that automatically extracts structured information from PDF resumes and recommends suitable job openings based on publicly available job search APIs. The system is based on natural language processing-based data extraction, stores the extracted information in a CSV database, and builds context-sensitive job queries. A basic web interface built on top of Flask supports PDF uploads and displays results in real-time. Our prototype is efficient for small to medium-sized datasets and recommends future work on incorporating intelligent ranking and resume optimization.

Introduction

Current job application processes are inefficient for both employers and applicants due to inconsistent resume formats and reliance on keyword-based resume screening, which can overlook qualified candidates. To address this, the paper proposes a lightweight, open-source, web-based system that automates resume parsing and job matching using natural language processing (NLP) techniques.

The system’s architecture includes:

• A Frontend Interface (HTML, CSS, Bootstrap) for easy PDF resume uploads and displaying job recommendations.

• A Resume Parsing Engine that uses pdfplumber and regex to extract structured data (experience, education, skills) from PDF resumes, handling varied formats.

• A CSV Storage System to save parsed data for querying and future use.

• A Query Generator that transforms parsed data into context-aware job search queries tailored to the applicant’s profile.

• Integration with SerpAPI, a job search API, to retrieve real-time, relevant job listings based on generated queries.

• A Job Display Module presenting results with filtering and sorting options.

• A Backend Server built on Flask managing data flow and user interactions.

• Security measures to protect user data and ensure privacy.

The methodology involves extracting and preprocessing PDF resume text, categorizing it via regex, storing structured data, generating personalized job search queries, and retrieving filtered job listings through SerpAPI. The system aims to improve job matching by moving beyond traditional keyword-based searches toward a more flexible, context-aware recommendation engine.

The implementation leverages Flask, pdfplumber, SerpAPI, CSV for storage, and Bootstrap for a responsive UI, focusing on scalability, ease of use, and accuracy in automated job matching.

Conclusion

This paper presented a lightweight, modular system for intelligent resume parsing and job recommendation using a web-based interface. The system effectively bridges the gap between unstructured resume formats and structured job search by extracting key candidate information such as experience, skills, and education, then using that information to formulate relevant job search queries. Our approach prioritizes simplicity and transparency while ensuring the system remains functional across various CV formats and job roles. Through the integration of pdfplumber for text extraction, pandas for CSV-based storage, and SerpAPI for dynamic job retrieval, we have demonstrated the potential for a real-time, interactive job recommendation engine that is cost-effective and easy to deploy. Evaluation metrics including parsing accuracy, query relevance, and job match precision reveal that the system performs reliably across diverse resumes, making it a promising tool for individuals, career counselors, and small HR departments. Looking forward, there are several directions for enhancing the system\'s functionality and scalability. Integrating a learning-to-rank model or a recommendation engine based on user feedback would significantly improve the relevance of suggested job listings. Additionally, supporting image-based resumes through OCR integration would allow the system to process a wider range of CV formats. Transitioning from CSV to a robust database such as PostgreSQL or MongoDB would improve performance with larger datasets and support user accounts for personalized dashboards. Another promising area is the implementation of a feedback loop where user interactions with job results (clicks, applications) are used to retrain the query generation module. Lastly, enabling resume optimization suggestions based on market trends and job descriptions could add further value, turning the platform into not just a search tool, but a comprehensive career enhancement assistant. These future improvements will further solidify the system’s applicability in real-world job-matching scenarios.

References

[1] Professor, V.V., Sushma, V., U.Deepak, Sai, P., &T.Mallikarjuna (2023). Improved Resume Parsing based on Contextual Meaning Extraction using BERT. 2023 7th International Conference on Intelligent Computing and Control Systems (ICICCS), 1702-1708. [2] Mohanty, S., Behera, A., Mishra, S., Alkhayyat, A., Gupta, D., & Sharma, V. (2023). Resumate: A Prototype to Enhance Recruitment Process with NLP based Resume Parsing. 2023 4th International Conference on Intelligent Engineering and Management (ICIEM), 1-6. [3] Liang, H. (2023). A study on the application of named entity recognition in resume parsing. Conference on Intelligent Computing and Human-Computer Interaction. [4] Manning, C.D., Surdeanu, M., Bauer, J., Finkel, J.R., Bethard, S., &McClosky, D. (2014). The Stanford CoreNLP Natural Language Processing Toolkit. Annual Meeting of the Association for Computational Linguistics. [5] Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., &Neubig, G. (2021). Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing. ACM Computing Surveys, 55, 1 - 35. [6] Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K., &Kuksa, P.P. (2011). Natural Language Processing (Almost) from Scratch. ArXiv, abs/1103.0398. [7] Prasad, B.L., Srividya, K., Kumar, K.N., Chandra, L.K., Dil, N.S., & Krishna, G.V. (2023). An Advanced Real-Time Job Recommendation System and Resume Analyser. 2023 International Conference on Self Sustainable Artificial Intelligence Systems (ICSSAS), 1039-1045. [8] Gadegaonkar, S., Lakhwani, D., Marwaha, S., &Salunke, P.A. (2023). Job Recommendation System using Machine Learning. 2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS), 596-603. [9] Shaikym, A., Zhalgassova, Z., &Sadyk, U. (2023). Design and Evaluation of a Personalized Job Recommendation System for Computer Science Students Using Hybrid Approach. 2023 17th International Conference on Electronics Computer and Computation (ICECCO), 1-7. [10] Yao, K., Pádua, G.B., Shang, W., Sporea, S., Toma, A., &Sajedi, S. (2018). Log4Perf: Suggesting Logging Locations for Web-based Systems\' Performance Monitoring. Proceedings of the 2018 ACM/SPEC International Conference on Performance Engineering. [11] Ismail, N.N., &Lokman, A.M. (2020). Kansei Engineering Implementation in Web-Based Systems: A Review Study. [12] Yao, K., de Pádua, G.B., Shang, W., Sporea, C., Toma, A., &Sajedi, S. (2019). Log4Perf: suggesting and updating logging locations for web-based systems’ performance monitoring. Empirical Software Engineering, 25, 488 - 531.

Copyright

Copyright © 2025 Kaustav Sanyal, Kinkini Gupta, Susmita Kuiry, Puspa Mahato, Bristi Dutta. 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.