Skip to content

Deep learning and data analysis repo, uncovering the latest neural network applications on blockchain attacks

Notifications You must be signed in to change notification settings

RektifyAI/adv-attack-analysis

Repository files navigation

A3: Advanced Attack Analysis

The Advanced Attack Analysis(A3) is a deep data analysis repo, uncovering the latest insights on attacks within the Web3 ecosystem. Methods such as EDA(exploratory data analysis), web scrapping, Python requests, vulnerability detection and Beneath API (halted acquired by RillzData) calls are used to retrieve attack specific data to explore what's happening on-chain with vulnerable Web3 products. The algorithm detects vulnerabilities through applied ANNs.

Rektify AI utilizes past attack data to perform prescriptive analytics using AI/ML as well as building models that will be used as data visualizations on the Rektify AI application. The system to curate live data will be done by analyzing on-chain and off-chain data from the blockchains of DeFi protocols we analyze.

Attacks recorded from then and now are utilized to make inferences so that a predictions based on data stored in the REAL API data warehouse can support analytical acrchitecture to detect high priority vulnerabilities before they lead to possible attacks. The use of ANNs support the vulnerability-attack prevention framework. The Rektify AI application will give alerts when attacks occur and warning messages when the deep learning system is triggered by vulnerabilities.

Index terms:
AI/ML - Artificial Intelligence and Machine Learning EDA - Exploratory Data Analysis (typically conducted to display relationships between metadata)
Supervised Learning - A machine learning technique that uses labeled data to train models to predict and classify outcomes
REAL - Rektify Exploratory Attack Library dataset
ANN - Artificial Neural Network
DApp - Decentralized Application


Applied Research: Discovering vulnerabilities before the attacks happen

Attacks only exist due to vulnerabilities and attack vectors that lead to possible intrusion. Throughout building the code base, within the attack analysis repo, we will extensively research specific vulnerabilities with the code for DeFi protocols.

The inception of DeFi is a contentious subject, but in this repo we aggregate attack data from the later since the inception of Bitcoin and Ethereum days. Around 2015, new chains began to emerge to partipate in the decentralized finance industry.


Registry

SWC Registry
DeFi Hack Labs
Ethernaut
Ethernaut CTF

About

Deep learning and data analysis repo, uncovering the latest neural network applications on blockchain attacks

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published