Etherum Smart Contracts Security improvement using machine learning techniques

Project Domain / Category
Machine Learning
Abstract / Introduction
Ethereum is a decentralized open source blockchain featuring smart contract functionality. Ether is
the native cryptocurrency token of the Ethereum platform. It is the second-largest cryptocurrency by
market capitalization, behind Bitcoin. Smart contracts are code files written in programming language
named solidity, as the block chain is based on smart contract, so the security of smart contract is very
important and vital to the security of blockchain system itself. Smart contracts can have many
vulnerabilities like recursive calls, unpredictable ether flows, Suspected input, transaction reordering
and insecure code patterns. Our goal in this project is to detect these vulnerabilities.

Functional Requirements:
1. The system must be able to analyse a complete solidity code file (smart contract).
2. System will use machine learning techniques to check for vulnerabilities in the smart contract
submitted.
3. System can check any 3 vulnerabilities from the given data set.
4. Deploying smart contract on ehterum network is not required.
The following flow can be followed to classify unsafe code (vulnerable code) and safe code of a smart
contract.