A Blockchain and Machine Learning Framework for Secure E-Voting Systems with Intrusion Detection

Abstract

Voting is a central component of a country’s political life cycle. Privacy, authentication, and integrity of citizens’ votes are essential requirements of any electronic voting programme. To address these concerns, this paper proposes a hybrid e-voting system that integrates personal and public blockchain with a machine learning–based intrusion detection mechanism. The personal blockchain governs voter registration and vote casting, while the public blockchain stores the Merkle root hash for result integrity verification. An ML-based intrusion detection system monitors voting data centres and e-voting stations for anomalous behaviour. Homomorphic encryption and zero-knowledge proofs preserve voter anonymity. Experimental evaluation demonstrates that the proposed framework achieves an accuracy of 97.4%, precision of 96.8%, recall of 97.1%, and F1-score of 96.9% in detecting intrusion attempts. The system also reduces transaction latency by 34% compared to conventional blockchain voting systems. Results confirm that the framework delivers transparency, tamper-resistance, and strong security guarantees, making it a viable solution for modern democratic elections.

Introduction

The text presents a secure blockchain-based electronic voting (e-voting) framework designed to address major weaknesses in existing voting systems, including centralized points of failure, DDoS attacks, vote tampering, weak authentication, and insider threats.

Proposed Solution

The system combines two blockchains with machine-learning intrusion detection:

• Permissioned personal blockchain: Uses Hyperledger Fabric and PBFT consensus to securely manage voter registration and encrypted ballots.
• Public blockchain: Stores Merkle root hashes and election results, allowing independent public verification without exposing individual votes.
• ML-based IDS: A Random Forest classifier monitors network traffic and detects suspicious behavior such as replay attacks, DDoS activity, and ballot stuffing.
• Merkle trees: Ensure vote integrity by making any alteration to an individual ballot detectable.
• Homomorphic encryption: Allows encrypted votes to be tallied without decrypting individual ballots.
• Zero-knowledge proofs: Help verify votes while preserving voter anonymity.
• Multi-factor authentication: Voters authenticate using credentials such as passwords and biometric tokens.

Key Contributions

The paper's main contributions are:

1. A hybrid personal–public blockchain architecture separating vote management from public result publication.
2. Integration of an ML-based intrusion detection system for real-time identification of malicious activity.
3. Merkle-tree verification for detecting vote manipulation.
4. Use of homomorphic encryption and zero-knowledge proofs to provide privacy and voter anonymity.
5. Experimental evaluation showing improvements in accuracy, latency, and scalability compared with existing approaches.

Related Work

Previous blockchain voting systems improved transparency and immutability but had limitations such as high Ethereum Proof-of-Work latency, centralized databases, inadequate integrity mechanisms, or high computational costs from encryption. The proposed system attempts to combine security, privacy, scalability, transparency, and intrusion detection in one framework.

Experimental Setup

The framework was tested using:

• KDD Cup 1999: 494,021 samples.
• UNSW-NB15: 257,673 samples.
• Simulated voting dataset: 50,000 legitimate and fraudulent voting transactions.

The system used Hyperledger Fabric, Ethereum/Ganache, Python, React, MongoDB, and machine-learning frameworks including Scikit-learn and TensorFlow.

Major Results

The Random Forest IDS achieved the best performance:

ML intrusion detection performance

Accuracy of the evaluated classifiers reported in the text.

0%30%60%90%120%Naïve BayesDecision TreeSVMNeural NetworkRandom ForestAccuracyAccuracy

• Random Forest achieved 97.4% accuracy, 96.8% precision, 97.1% recall, and 96.9% F1-score.
• The proposed system had a mean transaction latency of 1.82 seconds for 5,000 concurrent voters, compared with 2.76 seconds for the Ethereum PoW baseline.
• At 20,000 concurrent voters, latency increased to 3.14 seconds.
• The system reportedly sustained 820 transactions per second at 12 nodes, compared with 340 TPS for the single-blockchain configuration.
• Attack detection exceeded 95% across the reported attack categories.

Conclusion

This paper presented a hybrid blockchain and machine learning framework for secure, transparent, and privacy-preserving e-voting. The dual-blockchain architecture separates operational voting from public result verification, while an ML-based intrusion detection system provides real-time protection against cyberattacks on voting infrastructure. Merkle hash tree verification guarantees vote integrity, and homomorphic encryption combined with zero-knowledge proofs ensures ballot anonymity. Experimental results confirm 97.4% intrusion detection accuracy, a 34% reduction in transaction latency over PoW baselines, and linear scalability to 25,000 concurrent voters. The proposed system represents a significant advancement toward trustworthy digital democracy.

References

[1] A. B. Ayed, \"A conceptual secure blockchain-based electronic voting system,\" Int. J. Netw. Secur. Appl., vol. 9, no. 3, pp. 1–13, 2017. [2] Y. Sharma, P. Jain, and R. Nair, \"Blockchain-based secure online voting system,\" Int. J. Comput. Appl., vol. 185, no. 12, pp. 22–28, 2024. [3] A. Kumar, S. Verma, and A. Gupta, \"AI-driven voter authentication for online elections,\" Int. J. Inf. Secur. Appl., vol. 71, pp. 103–115, 2023. [4] P. Sharma and V. Patil, \"Smart voting system using cloud and IoT technologies,\" in Proc. ICETCE, 2024, pp. 45–52. [5] M. Brown and S. Lee, \"Enhancing online voting security with homomorphic encryption,\" J. Cryptograph. Res., vol. 11, no. 2, pp. 78–94, 2023. [6] D. Wilson and E. Carter, \"A comparative study of online voting systems: Challenges and future prospects,\" Int. J. Digit. Democracy, vol. 6, no. 1, pp. 14–31, 2024. [7] S. Nakamoto, \"Bitcoin: A peer-to-peer electronic cash system,\" Cryptography Mailing List, 2008. [Online]. Available: https://bitcoin.org/bitcoin.pdf [8] Hyperledger Foundation, \"Hyperledger Fabric 2.4 documentation,\" Linux Foundation, 2022. [Online]. Available: https://hyperledger-fabric.readthedocs.io [9] R. Cramer, R. Gennaro, and B. Schoenmakers, \"A secure and optimally efficient multi-authority election scheme,\" in Proc. EUROCRYPT, 1997, pp. 103–118. [10] M. Bellare and P. Rogaway, \"Random oracles are practical: A paradigm for designing efficient protocols,\" in Proc. ACM CCS, 1993, pp. 62–73. [11] V. Buterin, \"Ethereum: A next-generation smart contract and decentralised application platform,\" Ethereum White Paper, 2014. [Online]. Available: https://ethereum.org/whitepaper [12] A. Biryukov, D. Khovratovich, and I. Nikolic, \"Distinguisher and related-key attack on the full AES-256,\" in Proc. CRYPTO, 2009, pp. 231–249. [13] L. Lamport, R. Shostak, and M. Pease, \"The Byzantine generals problem,\" ACM Trans. Program. Lang. Syst., vol. 4, no. 3, pp. 382–401, 1982. [14] M. Castro and B. Liskov, \"Practical Byzantine fault tolerance,\" in Proc. OSDI, 1999, pp. 173–186. [15] R. C. Merkle, \"A digital signature based on a conventional encryption function,\" in Proc. CRYPTO, 1987, pp. 369–378. [16] P. Paillier, \"Public-key cryptosystems based on composite degree residuosity classes,\" in Proc. EUROCRYPT, 1999, pp. 223–238. [17] S. Goldwasser, S. Micali, and C. Rackoff, \"The knowledge complexity of interactive proof systems,\" SIAM J. Comput., vol. 18, no. 1, pp. 186–208, 1989. [18] N. Moustafa and J. Slay, \"UNSW-NB15: A comprehensive data set for network intrusion detection systems,\" in Proc. MilCIS, 2015, pp. 1–6. [19] KDD Cup 1999 Data. [Online]. Available: https://kdd.ics.uci.edu/databases/kddcup99/kddcup99.html [20] L. Breiman, \"Random forests,\" Mach. Learn., vol. 45, no. 1, pp. 5–32, 2001. [21] V. Cortier, D. Galindo, R. Küsters, J. Müller, and T. Truderung, \"SoK: Verifiability notions for e-voting protocols,\" in Proc. IEEE S&P, 2016, pp. 779–798. [22] B. Adida, \"Helios: Web-based open-audit voting,\" in Proc. USENIX Security, 2008, pp. 335–348. [23] P. McCorry, S. F. Shahandashti, and F. Hao, \"A smart contract for boardroom voting with maximum voter privacy,\" in Proc. FC, 2017, pp. 357–375. [24] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT Press, 2016. [25] S. Aggarwal and N. Kumar, \"Attacks on PKI and blockchain-based systems,\" J. Inf. Secur. Appl., vol. 58, pp. 102–119, 2021.

Copyright

Copyright © 2026 Mrs. A. Asrin Mahmootha, Dr. B. Aysha Banu, Mohammed Muhajir S, Mohamed Anas A, Mohamed Anas R. 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.