The blockchain is a rapidly growing field, and analyzing Ethereum transactions can provide valuable insights into transaction behaviors, gas fees, and patterns that may signal trends or anomalies. Our objective is to understand transaction behaviors on the Ethereum blockchain, particularly by identifying anomalies such as irregular gas fees or potential fraudulent activities. By examining transaction patterns from the last 20 blocks of Ethereum, this project aims to determine how transactional behaviors change over short periods and whether some patterns are not immediately visible.
Data will be collected through web scraping from the Etherscan block explorer at (https://etherscan.io/txs). Specifically, we will extract data from the last 20 Ethereum blocks, including transaction hashes, gas fees, sender and receiver addresses, and timestamps. This data is publicly available and can be gathered in real-time. Each block typically contains hundreds of transactions, so we expect to handle approximately 2,000 to 5,000 transactions for analysis. Given the nature of web scraping, this data will be organized in tables featuring various attributes, such as transaction hash, gas fees, and addresses.
This project will start by posing questions about Ethereum transactions. We will collect transaction data from the last 20 blocks using web scraping techniques. After gathering the data, we will analyze it to identify patterns and anomalies. The analysis will employ techniques such as clustering, anomaly detection, and trend analysis. The final deliverable will be an interactive visual dashboard that allows users to explore trends and correlations within the Ethereum blockchain data. Users will be able to filter the data by gas fees, block numbers, and transaction types.