Downloads: 2 | Views: 241 | Weekly Hits: ⮙2 | Monthly Hits: ⮙2
Research Paper | Information Technology | India | Volume 11 Issue 6, June 2022 | Popularity: 5.4 / 10
Enhancing Cloud-Based Smart Contract Security: A Hybrid AI and Optimization Approach for Vulnerability Prediction in FinTech
Ranadeep Reddy Palle, Haritha Yennapusa, Krishna Chaitanya Rao Kathala
Abstract: Financial industries operate within a framework of strict regulatory requirements, making compliance a top priority. Smart contracts, integral to the operations of FinTech companies, must align with these regulations. Cloud-based platform offers security as a service (SecaaS) to the scalable and cost-effective solution for analyzing, monitoring, and predicting vulnerabilities in smart contracts. This approach allows FinTech firms to concentrate on their core services while benefiting from specialized security tools. The potential consequences of smart contract vulnerabilities, such as financial losses, fraud, or data manipulation, underscore the critical need for proactive prediction and mitigation. By addressing vulnerabilities in advance, FinTech platforms can prevent financial losses and uphold the integrity of their transactions. Given that FinTech platforms handle customer funds, sensitive financial information, and automated transactions, maintaining trust and reliability is paramount. Predicting vulnerabilities plays a pivotal role in building and sustaining trust among users and stakeholders. This study introduces a hybrid artificial intelligence and optimization technique for smart contract vulnerability prediction in FinTech. The modified barnacles mating optimization (MBMO) algorithm is employed for the extraction of complex syntactic and semantic features, enhancing the accuracy of vulnerability predictions. Additionally, the general regressive artificial neural network (GR-ANN) is utilized to predict vulnerabilities, specifically describing vulnerability types in smart contracts deployed in a cloud environment. The evaluation of this framework involves rigorous testing using the ScrawID-real Ethereum smart contract benchmark dataset, demonstrating its capability and accuracy in predicting smart contract vulnerabilities. The study introduces a novel hybrid artificial intelligence and optimization technique aimed at predicting vulnerabilities in cloud-based smart contracts, specifically in the FinTech sector. Utilizing the modified barnacles mating optimization algorithm and the general regressive artificial neural network, this approach enhances the accuracy of vulnerability detection. The paper demonstrates the methods efficacy through rigorous testing with the ScrawID-real Ethereum smart contract benchmark dataset, highlighting its potential to bolster security in FinTech applications.
Keywords: smart contract, vulnerability prediction, cloud computing, artificial intelligence, machine learning, optimization technique
Edition: Volume 11 Issue 6, June 2022
Pages: 1959 - 1968
DOI: https://www.doi.org/10.21275/PR231222115735
Make Sure to Disable the Pop-Up Blocker of Web Browser