AI-Driven Career Intelligence and Job Fraud Detection Platform Using Machine Learning and Natural Language Processing

Abstract

The rapid proliferation of online recruitment platforms has significantly increased career exploration opportunities but has simultaneously exposed users to sophisticated employment scams. Traditional career guidance systems often provide static recommendations that fail to account for individual personality nuances or the security vulnerabilities present in the modern job market. This paper presents AegisPath, an AI-driven platform that integrates vocational synthesis with employment threat intelligence under a unified framework. The system utilizes an MBTI-based psychometric assessment engine to computationally derive personality types and generate personalized 6-month career roadmaps enriched with real-time India 2025 salary benchmarks. Simultaneously, it deploys an 8-signal Natural Language Processing (NLP) fraud detection engine that evaluates job postings for linguistic manipulation indicators, fee-extraction patterns, and URL legitimacy signals. The Career Intelligence Engine leverages a Large Language Model (LLM) backend with a Naive Bayes-grounded classification layer and content-based filtering to produce explainable, ranked career recommendations. Experimental evaluation on 140 student respondents demonstrated an overall recommendation accuracy of 88.6% and a fraud detection precision of 91.3%, with user satisfaction exceeding 86% across coverage, novelty, and diversity metrics. AegisPath is the first platform to simultaneously address the Guidance Gap and the Security Gap in online vocational systems.

Introduction

The text discusses how online recruitment systems have created both opportunities and challenges for students entering the job market, highlighting two major gaps: lack of personalized career guidance (“Guidance Gap”) and rising employment fraud (“Security Gap”). To address these issues, the authors propose AegisPath, an AI-based career guidance platform.

AegisPath integrates multiple components: psychometric profiling using a simplified MBTI test, a hybrid career recommendation engine combining Naive Bayes classification, skill matching, and collaborative filtering, and a learning roadmap generator that calculates skill gaps through an Occupational Readiness Score (ORS). It also includes an 8-signal NLP-based fraud detection system to identify suspicious job postings and a conversational CareerBot that explains recommendations and provides guidance.

The system is built using a full-stack MERN architecture and evaluated on 140 students and 2,400 job postings. Results show that AegisPath outperforms baseline models, achieving 88.6% recommendation accuracy and 90.4% fraud detection accuracy, along with high user satisfaction.

Conclusion

This paper presented AegisPath, an AI-driven platform that simultaneously addresses the Guidance Gap and the Security Gap in online career exploration. By integrating MBTI-based psychometric profiling, a hybrid Naive Bayes and content-based recommendation engine, an 8-signal NLP fraud detection module, and a LLM-powered CareerBot, AegisPath provides students with a secure, personalized, and explainable career planning environment. The system\'s recommendation engine builds upon and extends the EDM-GT-grounded p-NB framework of Siswipraptini et al. [4] by incorporating real-time job market data, skill-gap quantification, and ORS-driven roadmap generation. Experimental evaluation on 140 students demonstrated 88.6% recommendation accuracy and 91.3% fraud detection precision, with user satisfaction exceeding 86% across all measured dimensions. Future work will pursue: (1) Deep learning integration—exploring LSTM-based sequence models for temporal fraud pattern detection and transformer-based embeddings for richer semantic job matching; (2) Expanded geographic coverage—extending salary benchmarking and job taxonomy beyond India to support international student populations; (3) Longitudinal validation—a 12-month follow-up study tracking career outcomes of AegisPath users to assess long-term roadmap recommendation quality.

References

REFERENCES [1] D. B. Guruge, R. Kadel, and S. J. Halder, \"The state of the art in methodologies of course recommender systems—A review of recent research,\" Data, vol. 6, no. 2, pp. 1–30, Feb. 2021. [2] Ministry of Home Affairs, Republic of India, \"Annual Report on Cybercrime Statistics,\" New Delhi, 2024. [3] K. A. Hoff, Q. C. Song, C. J. M. Wee, W. M. J. Phan, and J. Rounds, \"Interest fit and job satisfaction: A systematic review and meta-analysis,\" J. Vocational Behav., vol. 123, Dec. 2020, Art. no. 103503. [4] P. C. Siswipraptini, H. L. H. S. Warnars, A. Ramadhan, and W. Budiharto, \"Personalized career-path recommendation model for information technology students in Indonesia,\" IEEE Access, vol. 12, pp. 49092–49105, 2024. [5] M. E. Ibrahim, Y. Yang, D. L. Ndzi, G. Yang, and M. Al-Maliki, \"Ontology-based personalized course recommendation framework,\" IEEE Access, vol. 7, pp. 5180–5199, 2019. [6] N. T. Son, J. Jaafar, I. A. Aziz, and B. N. Anh, \"Meta-heuristic algorithms for learning path recommender at MOOC,\" IEEE Access, vol. 9, pp. 59093–59107, 2021. [7] D. Mhamdi, R. Moulouki, M. Y. El Ghoumari, M. Azzouazi, and L. Moussaid, \"Job recommendation based on job profile clustering and job seeker behavior,\" Proc. Comput. Sci., vol. 175, pp. 695–699, Jan. 2020. [8] L. D. Kumalasari and A. Susanto, \"Recommendation system of IT jobs using collaborative filtering based on LinkedIn skills endorsement,\" SISFORMA, vol. 6, no. 2, pp. 63–72, Feb. 2020. [9] S. Bharadwaj, S. Sridhar, R. Choudhary, and R. Srinath, \"Persona traits identification based on Myers-Briggs type indicator—A text classification approach,\" in Proc. ICACCI, Sep. 2018, pp. 1076–1082. [10] K. A. Hoff and J. Rounds, \"Interest congruence and performance: Revisiting recent meta-analytic findings,\" J. Vocational Behav., vol. 98, pp. 138–151, Feb. 2017. [11] B. E. V. Comendador, W. F. C. Becbec, and J. R. P. D. Guzman, \"Implementation of fuzzy logic technique for choosing appropriate career path,\" Int. J. Mach. Learn. Comput., vol. 10, no. 2, pp. 339–345, Feb. 2020. [12] R. Mishra and S. Rathi, \"Enhanced DSSM technique for job recommendation,\" J. King Saud Univ. Comput. Inf. Sci., vol. 34, no. 9, pp. 7790–7802, Oct. 2022. [13] S. Jadhav, A. More, R. Patil, and S. Kadam, \"Network intrusion and fraud job detection using machine learning,\" IJSRCSEIT, vol. 8, no. 3, pp. 208–216, 2022. [14] E. Zangerle and C. Bauer, \"Evaluating recommender systems: Survey and framework,\" ACM Comput. Surv., vol. 55, no. 8, pp. 1–38, Aug. 2023.

Copyright

Copyright © 2026 NathishKumar S, Sanjai G, Nithivananthan M, Pasupathi I, Ms. Rakshanadevi S. 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.