Blockchain Retranattack Identification and Detection Via Association Learning

The clarity of Blockchain and Smart Contracts revolutionized the concept of financial services in 2024.In contrast, as per Forta.org,and Dirty Block reports saw sophisticated attacks like reentrancy attacks, where attackers successfully exploited system calls that caused a loss of revenue and created question marks on Blockchain-based applications.To guard against this challenge, we propose a Retransattack identification and detection Algorithm which successfully finds five types of attacks, with the help of an Association learning Retransattack identification and detection Algorithm to identify these sorts of attacks transfer pattern, safe math, self-destructive, send pattern and timestamp block hash method with the help of Association techniques e.g.pattern relationship on dataset 3,000 smart contract transactions of 2023 and 2024.To summarize, we will conclude our Finding with a Medium-scale Ethereum smart contract transactions attack detection of 500 days, and we will convey the total number of reentrancy attacks and similar attacks and their transaction behaviour with the help of association learning resulting in revenue losses in millions.This research contribution is one medium-sized dataset because we collected it and applied Exploratory Data Analysis on the dataset from 20 different sites e.g. from BNB, ether scan and cryptoscan.In Future, we will work on the Accuracy, precision and F1-score of the dataset and optimize hyperparameters to improve model performance.We will test and evaluate model deployment to monitor whether it applies to production environments.