Smart Contract Vulnerability Detection Method Based on Feature Graph and Multiple Attention Mechanisms
作者机构：School of Cyber SecurityGansu University of Political Science and LawLanzhou730070China
出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)
年 卷 期：2024年第79卷第5期
页 面：3023-3045页
核心收录：
学科分类：07[理学] 0701[理学-数学] 070101[理学-基础数学]
主 题：Blockchain smart contracts deep learning graph neural networks
摘 要：The fast-paced development of blockchain technology is ***,the security concerns of smart contracts represent a significant challenge to the stability and dependability of the entire blockchain *** smart contract vulnerability detection primarily relies on static analysis tools,which are less efficient and *** deep learning methods have improved detection efficiency,they are unable to fully utilize the static relationships within ***,we have adopted the advantages of the above two methods,combining feature extraction mode of tools with deep learning ***,we have constructed corresponding feature extraction mode for different vulnerabilities,which are used to extract feature graphs from the source code of smart ***,the node features in feature graphs are fed into a graph convolutional neural network for training,and the edge features are processed using a method that combines attentionmechanismwith gated ***,the revised node features and edge features are concatenated through amulti-head *** result of the splicing is a global representation of the entire feature *** method was tested on three types of data:Timestamp vulnerabilities,reentrancy vulnerabilities,and access control vulnerabilities,where the F1 score of our method reaches 84.63%,92.55%,and 61.36%.The results indicate that our method surpasses most others in detecting smart contract vulnerabilities.
维普期刊数据库