Automated Job Offer Trust Scoring System

Abstract

The Automated Job Offer Trust Scoring System (AJOTSS) is developed to help job seekers evaluate the legitimacy of job-related emails and recruitment offers. Scammers often use fake domains, copied company branding, and payment requests to deceive users. The proposed system analyses job-related email content and performs multiple verification checks, including domain name validation, NLP-based content scanning, fraud database lookup, and inspection of embedded links for suspicious payment or redirection behaviour. Based on these verification constraints, the system calculates a Trust Score between 0 and 100 that represents the credibility of the job offer. AJOTSS provides a practical, rule-based solution to reduce recruitment fraud risk and assist job seekers in making informed decisions.

Introduction

The Automated Job Trust Scoring System (AJOTSS) is a web-based application developed to help job seekers identify fraudulent recruitment emails. With the rapid growth of online recruitment platforms, fake job offers have become increasingly sophisticated, making manual verification difficult and unreliable. AJOTSS addresses this problem by automatically analyzing job-related emails and generating a Trust Score (0–100) that indicates their authenticity and reliability.

The system uses a multi-level verification approach consisting of four independent analysis modules: Domain Verification, NLP-Based Content Analysis, Fraud Database Checking, and Link Inspection. Domain verification checks whether the sender's email domain matches an official company domain. Content analysis uses Natural Language Processing (NLP) to detect suspicious keywords related to urgency, payment requests, training fees, or requests for personal information. The fraud database module compares company names and domains against a database of known fraudulent recruiters, while the link inspection module evaluates URLs for suspicious characteristics such as shortened links, payment-related pages, credential-harvesting attempts, and unusual URL structures.

The system is implemented using Python and Flask, with a simple HTML/CSS-based web interface that allows users to paste job emails and receive instant results. A rule-based Trust Score Engine combines the risk values generated by each module and classifies emails into categories such as TRUSTED, SAFE, CAUTION, SUSPICIOUS, or FRAUDULENT.

Testing results demonstrated that the system effectively distinguishes genuine recruitment emails from fraudulent ones. Emails containing payment-related keywords are immediately classified as fraudulent with a score of zero, while emails from verified official company domains receive the highest trust score. The modular architecture ensures scalability, efficiency, and ease of maintenance while providing users with a fast, user-friendly, and reliable method for evaluating recruitment emails and reducing the risk of job-related scams.

Conclusion

The Automated Job Offer Trust Scoring System provides an effective, rule-based solution for identifying fraudulent job emails. By combining four independent verification modules domain verification, NLP content analysis, fraud database lookup, and link inspectioninto a weighted trust score, AJOTSS enables job seekers to make informed decisions quickly. The system successfully demonstrated the ability to detect payment demands, impersonated domains, and fraudulent recruiters, while correctly validating genuine communications from verified corporate domains.

References

[1] S. Chawla, J. Cichy, and T. G. Papaioannou, \"Automatic Detection of Online Recruitment Frauds: Characteristics, Methods, and a Public Dataset,\" Future Internet (MDPI), vol. 9, no. 6, pp. 1–12, 2017. [2] M.Khonji, Y. Iraqi, and A. Jones, \"Phishing Detection: A Literature Survey,\" IEEE Communications Surveys and Tutorials, vol. 15, no. 4, pp. 2091–2121, 2013. [3] M. A. U. Masud, T. Al-Khateeb, L. Khan, B. Thuraisingham, and K. W. Hamlen, \"Detecting Phishing Websites Using Machine Learning Techniques,\" in Proc. IEEE Int. Conf. on Information Reuse and Integration, 2008, pp. 1–6. [4] D. Oliveira, H. Rocha, and H. Yang, \"Dissecting Spear Phishing Emails for Older vs Young Adults,\" in Proc. ACM CHI Conf. on Human Factors in Computing Systems, 2017, pp. 6412–6424. [5] A.Josang, R. Ismail, and C. Boyd, \"A Survey of Trust and Reputation Systems for Online Service Provision,\" Decision Support Systems (Elsevier), vol. 43, no. 2, pp. 618–644, 2007.

Copyright

Copyright © 2026 Pavithra K, Poojalakshmi G, Naziya Banu I, Serin S, Sowndharya M. 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.