This post presents the motivation and core technology of Onchain Reputation System, with extensive applications: anti-Sybils and effective airdrops, user qualification and segmentation, future user-value prediction, onchain trust profile.

Onchain Reputation matters

Onchain Reputation is a key problem in Coinbase’s Request for Builders, which assigns a reputation score to each public address based on the graph of onchain transaction data. It is expected to be applied in loan-credit (like a decentralized FICO score), merchant ratings, fraud and Sybil prevention, airdrops to right persons.

In the wishlist of Q3.2024 hackathon, BNB Chain raised a critical need of anti-Sybil tools to distinguish genuine users from bots in the rapidly growing Web3Gamefi.

Core technology

Onchain Reputation Ranking is an innovative system, based on Google’s Pagerank, a trillion-dollar algorithm widely used in Web search engine, recommendation, social rating, academic ranking, etc. Roughly, PageRank works by counting the number and quality of links to a page to determine a rough estimate of how important the website is. The underlying assumption is that more important websites are likely to receive more links from other websites in the Webgraph.

Besides extensive applications in Web2, the idea of Pagerank can be extended to any social graphs (consisting of nodes and edges/links). For example, on blockchain, if we consider public addresses (individual-owned accounts or contracts) as nodes (like web-pages) and transactions (e.g. token transfers, contract interactions) as edges/links, then we can apply Pagerank to measure the importance or influence of an account within the blockchain ecosystem. The onchain accounts may represent individuals or applications operation onchain. The key purpose is to extract social insights and user behaviors from onchain transactional graphs. In the following, we propose the intuitions of the Onchain Reputation Ranking.

• The reputation score of a public address (or account) within a blockchain is derived from its Pagerank score, computed from the transaction graph constructed on the blockchain data.

• More IN-transaction or Indegree (receiving value), then higher reputation score.

• IN-transaction from a high reputation account result in greater reputation score than receiving from a low reputation account.

• OUT-transaction or Outdegree (sending value) result in a lower reputation score.

• Weights are applied: account & transaction age, transaction value, diversity, etc. For instance, latest transactions are more meaningful than the old ones, greater value transactions are better.

• It should care Sybil/spam attacks, manipulation, sinkage effect.

Onchain Reputation problem and onchain-adaptive Pagerank algorithms have been investigated since 2019, with 3 published research papers on ACM, IGI and IEEE, notable scientific journals in computer science and software disciplines.

Applications of Onchain Reputation

The aforementioned researches on IGI and IEEE, proposed a wide application perspectives of Onchain Reputation:

• Evaluating impact and quality of dapps, Web3 applications in various sectors, then suggesting for investors and traders.

• Qualifying Web3 users and classifying them according to various levels and categories (e.g. DeFi, NFT, gamefi, socialfi, meme, etc). This is an useful case in targeted marketing onchain.

• Mitigating Sybil attacks and spamming transactions onchain.

• Forecasting future user value, hence offering an effective measurement/metrics for airdrops or marketing campaign in Web3.

• Establishing trust and credibility in Web3, facilitating real users to build their trustable profile onchain with anonymity and get financial rewards/benefits fairly.

In particular, three meaningful use-cases of Onchain Reputation score validated by Arbitrum, Optimism and Layerzero airdrops:

• Classify eligible wallets with high accuracy (AUC 0.82)

• Forecast future user-value (captured 92.4% on Optimism, 74.8% on Arbitrum)

• Captured most Sybils (72% Sybil addresses of Layerzero airdrop)

Forecasting future user-value is very important. All marketing campaigns and airdrops are eventually to target high user-value generation in the future.

In essence, onchain-extracted insights and user behaviors quantified by Onchain Reputation score will give trustable metrics and data-driven actions for Web3 builders, investors and users.

Engage to get more financial rewards with us

Website, Twitter, Telegram and subscribe to earn more airdrops in the future