Home / Current Issue / Paper 1716368
VeriTrust: An AI-Powered Decentralized Reputation System for the Gig Economy
Subject area: Science,Engineering and Technology · Area of research: Computer Science
DOI: 10.64388/IREV9I10-1716368
Abstract
Trust in the gig economy lives and dies by reputation. Platforms like Upwork and Fiverr use review scores as stand-ins for credibility, yet the systems underlying those scores are opaque, siloed, and surprisingly easy to manipulate. This paper presents VeriTrust, a fully working AI-powered reputation system that pairs blockchain immutability with natural language processing to give freelancers a tamper-proof, portable reputation they genuinely own. At its core, VeriTrust runs an application-level Proof-of-Work blockchain with SHA-256 mining and RSA-2048 cryptographic signing so that every review is verifiably authentic. Two Solidity smart contracts are deployed on Polygon Amoy—one storing full data for maximum auditability, and a leaner VeriTrustLite variant that trims gas costs by roughly 80% through struct packing, custom errors, and hash-only storage. A three-step prepare/confirm write protocol ties each on-chain anchor to the matching PostgreSQL record, and reputation is keyed on RSA public keys rather than wallet addresses, so users can carry their history across chains and platforms without asking anyone’s permission. A React.js DApp with MetaMask integration, a live blockchain explorer, and a chain indexer round out the implementation. Smart contract test coverage is 100% across all statement branches.
Keywords
Blockchain, Decentralized Application, Gig Economy, Natural Language Processing, Proof-Of-Work, Reputation Systems, RSA Cryptography, Sentiment Analysis, Smart Contracts.
References
[1] C. Hillebrand and A. P. Coetzee, “Towards Reputation-as-a-Service in the Cloud,” in Proc. Int. Conf. Cloud Computing and Services Science (CLOSER), 2013, pp. 371–380.
[2] D. C. Nguyen, P. N. Pathirana, M. Ding, and A. Seneviratne, “Blockchain-based Trust System for Decentralised Applications: A Survey,” IEEE Access, vol. 9, pp. 118270–118299, 2021.
[3] Z. Zheng, S. Xie, H. Dai, X. Chen, and H. Wang, “Blockchain-Based Decentralized Application: A Survey,” IEEE Trans. Syst., Man, Cybern.: Syst., vol. 53, no. 6, pp. 3725–3743, 2023.
[4] M. M. Hassan and M. S. Islam, “Impact of Sentiment Analysis in Fake Online Review Detection,” in Proc. Int. Conf. ICT for Sustainable Development (ICT4SD), 2021, pp. 673–683.
[5] S. Kotha, “Distributed Fake Review Detection and Real-Time Anomaly Detection Using Machine Learning,” J. Computational Intelligence, vol. 12, no. 1, pp. 45–62, 2025.
[6] D. Yu, Y. Zhang, Z. Huang, W. Wang, and P. S. Yu, “Graph Learning for Fake Review Detection,” in Proc. ACM Web Conf. (WWW), 2022, pp. 2879–2888.
[7] D. Martens and W. Maalej, “ReviewChain: Untampered Product Reviews on the Blockchain,” in Proc. IEEE Int. Conf. Software Architecture Companion (ICSA-C), 2018, pp. 126–129.
[8] M. Bhatia, A. Kumar, and D. Sangwan, “WorkerRep: Immutable Reputation System For Crowdsourcing Platform Using Blockchain,” in Proc. Int. Conf. Communication Systems & Networks (COMSNETS), 2020, pp. 748–753.
[9] C. Toxtli, A. Suri, and S. Savage, “Reputation Agent: Prompting Fair Reviews in Gig Markets,” in Proc. The Web Conf. (WWW), 2020, pp. 1343–1349.
[10] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,” in Proc. NAACL-HLT, 2019.
[11] S. Nakamoto, “Bitcoin: A Peer-to-Peer Electronic Cash System,” 2008. [Online]. Available: https://bitcoin.org/bitcoin.pdf
[12] V. Buterin, “A Next-Generation Smart Contract and Decentralized Application Platform,” Ethereum Whitepaper, 2014. [Online]. Available: https://ethereum.org
How to cite this paper
@article{1716368,
author = {Muiz Zatam, Angad Muthyala, Sarvesh Varvatkar, Jaspreet Kaur},
title = {VeriTrust: An AI-Powered Decentralized Reputation System for the Gig Economy},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {10},
pages = {1687-1691},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1716368.pdf},
abstract = {Trust in the gig economy lives and dies by reputation. Platforms like Upwork and Fiverr use review scores as stand-ins for credibility, yet the systems underlying those scores are opaque, siloed, and surprisingly easy to manipulate. This paper presents VeriTrust, a fully working AI-powered reputation system that pairs blockchain immutability with natural language processing to give freelancers a tamper-proof, portable reputation they genuinely own. At its core, VeriTrust runs an application-level Proof-of-Work blockchain with SHA-256 mining and RSA-2048 cryptographic signing so that every review is verifiably authentic. Two Solidity smart contracts are deployed on Polygon Amoy—one storing full data for maximum auditability, and a leaner VeriTrustLite variant that trims gas costs by roughly 80% through struct packing, custom errors, and hash-only storage. A three-step prepare/confirm write protocol ties each on-chain anchor to the matching PostgreSQL record, and reputation is keyed on RSA public keys rather than wallet addresses, so users can carry their history across chains and platforms without asking anyone’s permission. A React.js DApp with MetaMask integration, a live blockchain explorer, and a chain indexer round out the implementation. Smart contract test coverage is 100% across all statement branches.},
keywords = {Blockchain, Decentralized Application, Gig Economy, Natural Language Processing, Proof-Of-Work, Reputation Systems, RSA Cryptography, Sentiment Analysis, Smart Contracts.},
month = {April},
doi = {https://doi.org/10.64388/IREV9I10-1716368}
}