Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset
With the advent of blockchain networks, there has been a transition from traditional contracts to Smart Contracts (SCs), which are crucial for maintaining trust within these networks. Previous methods for analyzing SCs vulnerabilities typically suffer from a lack of accuracy and effectiveness. Many...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-06-01
|
| Series: | Blockchain: Research and Applications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2096720924000666 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849419155093061632 |
|---|---|
| author | Sepideh HajiHosseinKhani Arash Habibi Lashkari Ali Mizani Oskui |
| author_facet | Sepideh HajiHosseinKhani Arash Habibi Lashkari Ali Mizani Oskui |
| author_sort | Sepideh HajiHosseinKhani |
| collection | DOAJ |
| description | With the advent of blockchain networks, there has been a transition from traditional contracts to Smart Contracts (SCs), which are crucial for maintaining trust within these networks. Previous methods for analyzing SCs vulnerabilities typically suffer from a lack of accuracy and effectiveness. Many of them, such as rule-based methods, machine learning techniques, and neural networks, also struggle to detect complex vulnerabilities due to limited data availability. This study introduces a novel approach to detecting, identifying, and profiling SC vulnerabilities, comprising two key components: an updated analyzer named SCsVulLyzer (V2.0) and an advanced Genetic Algorithm (GA) profiling method. The analyzer extracts 240 features across different categories, while the enhanced GA, explicitly designed for profiling SC vulnerabilities, employs techniques such as penalty fitness function, retention of elites, and adaptive mutation rate to create a detailed profile for each vulnerability. Furthermore, due to the lack of comprehensive validation and evaluation datasets with sufficient samples and diverse vulnerabilities, this work introduces a new dataset named BCCC-SCsVul-2024. This dataset consists of 111,897 Solidity source code samples, ensuring the practical validation of the proposed approach. Additionally, three types of taxonomies are established, covering SC literature review, profiling techniques, and feature extraction. These taxonomies offer a systematic classification and analysis of information, enhancing the efficiency of the proposed profiling technique. Our proposed approach demonstrated superior capabilities with higher precision and accuracy through rigorous testing and experimentation. It not only showed excellent results for evaluation parameters but also proved highly efficient in terms of time and space complexity. Moreover, the concept of the profiling technique makes our model highly transparent and explainable. These promising results highlight the potential of GA-based profiling to improve the detection and identification of SC vulnerabilities, contributing to enhanced security in blockchain networks. |
| format | Article |
| id | doaj-art-4bf6a3fc11f44bfc84ef951a467a34f5 |
| institution | Kabale University |
| issn | 2666-9536 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Blockchain: Research and Applications |
| spelling | doaj-art-4bf6a3fc11f44bfc84ef951a467a34f52025-08-20T03:32:12ZengElsevierBlockchain: Research and Applications2666-95362025-06-016210025310.1016/j.bcra.2024.100253Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark datasetSepideh HajiHosseinKhani0Arash Habibi Lashkari1Ali Mizani Oskui2Computer Science, Lassonde School of Engineering, York University, Toronto M3J 1P3, Ontario, Canada; Behavior-Centric Cybersecurity Center (BCCC), School of Information Technology, York University, Toronto M3J 1P3, Ontario, Canada; Corresponding author.Computer Science, Lassonde School of Engineering, York University, Toronto M3J 1P3, Ontario, Canada; Behavior-Centric Cybersecurity Center (BCCC), School of Information Technology, York University, Toronto M3J 1P3, Ontario, CanadaFinancial and Crypto Advisory of Switzerland (FiCAS AG), Zug 6300, SwitzerlandWith the advent of blockchain networks, there has been a transition from traditional contracts to Smart Contracts (SCs), which are crucial for maintaining trust within these networks. Previous methods for analyzing SCs vulnerabilities typically suffer from a lack of accuracy and effectiveness. Many of them, such as rule-based methods, machine learning techniques, and neural networks, also struggle to detect complex vulnerabilities due to limited data availability. This study introduces a novel approach to detecting, identifying, and profiling SC vulnerabilities, comprising two key components: an updated analyzer named SCsVulLyzer (V2.0) and an advanced Genetic Algorithm (GA) profiling method. The analyzer extracts 240 features across different categories, while the enhanced GA, explicitly designed for profiling SC vulnerabilities, employs techniques such as penalty fitness function, retention of elites, and adaptive mutation rate to create a detailed profile for each vulnerability. Furthermore, due to the lack of comprehensive validation and evaluation datasets with sufficient samples and diverse vulnerabilities, this work introduces a new dataset named BCCC-SCsVul-2024. This dataset consists of 111,897 Solidity source code samples, ensuring the practical validation of the proposed approach. Additionally, three types of taxonomies are established, covering SC literature review, profiling techniques, and feature extraction. These taxonomies offer a systematic classification and analysis of information, enhancing the efficiency of the proposed profiling technique. Our proposed approach demonstrated superior capabilities with higher precision and accuracy through rigorous testing and experimentation. It not only showed excellent results for evaluation parameters but also proved highly efficient in terms of time and space complexity. Moreover, the concept of the profiling technique makes our model highly transparent and explainable. These promising results highlight the potential of GA-based profiling to improve the detection and identification of SC vulnerabilities, contributing to enhanced security in blockchain networks.http://www.sciencedirect.com/science/article/pii/S2096720924000666Smart contracts (SCs)VulnerabilityVulnerable smart contractsVulnerability profilingGenetic algorithm |
| spellingShingle | Sepideh HajiHosseinKhani Arash Habibi Lashkari Ali Mizani Oskui Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset Blockchain: Research and Applications Smart contracts (SCs) Vulnerability Vulnerable smart contracts Vulnerability profiling Genetic algorithm |
| title | Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset |
| title_full | Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset |
| title_fullStr | Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset |
| title_full_unstemmed | Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset |
| title_short | Unveiling smart contract vulnerabilities: Toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset |
| title_sort | unveiling smart contract vulnerabilities toward profiling smart contract vulnerabilities using enhanced genetic algorithm and generating benchmark dataset |
| topic | Smart contracts (SCs) Vulnerability Vulnerable smart contracts Vulnerability profiling Genetic algorithm |
| url | http://www.sciencedirect.com/science/article/pii/S2096720924000666 |
| work_keys_str_mv | AT sepidehhajihosseinkhani unveilingsmartcontractvulnerabilitiestowardprofilingsmartcontractvulnerabilitiesusingenhancedgeneticalgorithmandgeneratingbenchmarkdataset AT arashhabibilashkari unveilingsmartcontractvulnerabilitiestowardprofilingsmartcontractvulnerabilitiesusingenhancedgeneticalgorithmandgeneratingbenchmarkdataset AT alimizanioskui unveilingsmartcontractvulnerabilitiestowardprofilingsmartcontractvulnerabilitiesusingenhancedgeneticalgorithmandgeneratingbenchmarkdataset |