Detecting the vulnerabilities of smart contracts with machine learning method

نوع: Type: thesis

مقطع: Segment: masters

عنوان: Title: Detecting the vulnerabilities of smart contracts with machine learning method

ارائه دهنده: Provider: maryam hemati azandaryani

اساتید راهنما: Supervisors: Dr. Mehdi Sakhaei Nia

اساتید مشاور: Advisory Professors:

اساتید ممتحن یا داور: Examining professors or referees: Dr. Shakoor Vakilian and Dr. Reza Mohammadi

زمان و تاریخ ارائه: Time and date of presentation: 2024

مکان ارائه: Place of presentation: Faculty of Engineering

چکیده: Abstract: Smart contracts are decentralized programs, the code of which can implement any algorithm; And usually facilitates the exchange of money, assets, stocks, etc. These smart contracts have digital currencies. By executing the contract, these valuable assets can be stored, manipulated, managed and moved, which can lead to security breaches, and these breaches can lead to Huge financial losses and destroy the stability of the blockchain. Research has been done in the field of smart contract security. Smart contract analysis methods are based on static analysis, dynamic analysis and formal validation. There is currently no tool that can identify all the weaknesses of smart contracts, but a combination of multiple tools can identify current contract vulnerabilities. But if the vulnerability is not predefined and the vulnerability is new, it can not be identified by the tools available in these three methods. In addition, most existing tools have a high time overhead, require a lot of resources or a lot of complexity. This dissertation tries to identify vulnerabilities that are not seen in smart contracts or reduce the overhead time of vulnerability identification by using machine learning methods or combining machine learning methods with traditional and traditional analysis methods.

فایل: ّFile: Download فایل