Titulo Estágio
Analysis of distributed ledger transaction networks
Áreas de especialidade
Engenharia de Software
Local do Estágio
DEI
Enquadramento
Blockchain networks maintain publicly available ledgers of transactions. Due to the storage constraints, nodes do not index data stored in the blocks. Given the massive growth in Ethereum based protocols such as NFT (non-fungible token) marketplaces, the need to detect potential anomalies, unethical behaviour and manipulation is of great importance. To achieve this, two things are needed, an indexed data structure fit for such analysis, and a heuristic algorithm to analyse large transaction graphs in order to detect potential anomalies.
Transactions between individuals form a directed graph in which wallets are represented as vertexes, and transactions as edges. However, unlike traditional approaches, blockchain networks are pseudonymous and hence a user can own an arbitrary number of wallets. Identifying a set of wallets belonging to the same person can be done using heuristics. Of particular importance is the ability to detect cycles of various lengths, which include a given set of wallets.
To achieve this, the large graph needs to be properly constructed to allow fast cycle identification. Ethereum-etl is a command-line interface software that can be used to parse the Ethereum blockchain using a local fully synced node. After obtaining structured data, a transaction network must be constructed. Moreover, transactions, which are essentially smart contract calls must be filtered from the base layer Ethereum transactions.
The NFT marketplace is a set of smart contracts deployed on the Ethereum network. Actors create(mint) NFTs and offer them for sale using the decentralized application (dApp) such as OpenSea. However, with the massive influx in the value of the entire NFT marketplace, it is important to ask whether these valuations are a product of the traditional supply and demand function, or manipulated.
One naive approach to inflate the valuation of an NFT would be as follows:
* User U, uses wallet A to mint an NFT.
* User U, lists NFT for sale at a given price.
* User U uses wallet B to buy bid on the NFT inflating its price
Since both wallets are owned by A, the cost of inflating the price is only the gas fees spent for purchasing the NFT. The main idea is search the transaction graph to identify wallets that likely belong to the same person by analysing the edges between wallets. To achieve this, two graphs must be constructed. One is the graph derived from selling and buying NFTs on a decentralized marketplace, and the other is a graph of base layer transactions.
To goal is then to construct these graphs, enable efficient storage and querying, and finally implement algorithms for searching cycles in both graphs, and matching the results.
Objetivo
The ultimate goal is to create a tool that enables analysis of the Ethereum blockchain and perform some steps of analysis on the blockchain. This involves a number of possible steps:
* How to efficiently store and index large transaction graphs for querying?
* Can identifying cycles of various lengths be used to link a subset of wallets to an identity?
* Are NFT decentralized marketplaces artificially inflating the price?
Plano de Trabalhos - Semestre 1
- Study the state of the art (2 months).
- Define the application requirements (1 month).
- Start implementation (1 month).
- Write intermediate report (1 month).
Plano de Trabalhos - Semestre 2
- Implement the application (2 months).
- Collect data and perform analysis (1 month).
- Write scientific paper (1 month).
- Write final report (1 month).
Condições
The work should take place at the Centre for Informatics and Systems of the University of Coimbra (CISUC) at the Department of Informatics Engineering of the University of Coimbra. The work will occur in close cooperation with a research group from the University of Primorska, Slovenia, which is leading this research effort.
An 875,98 euros per month scholarship is foreseen for 3 months. The attribution of the scholarship is subject to a public application.
Orientador
Filipe Araujo, Aleksandar Tošić, Paulo Rupino
filipius@uc.pt 📩