# Nomis - Web3 Rating Score

**Published by:** [unchase](https://paragraph.com/@unchase/)
**Published on:** 2023-01-15
**URL:** https://paragraph.com/@unchase/nomis-web3-rating-score

## Content

Nomis ProtocolNomis Protocol is a web3 credential and scoring data network. Built on an open and collaborative infrastructure, the protocol helps web3 developers and other protocols leverage credential data to build better products and communities. At the same time, data curators are rewarded when credentials are used in Nomis Protocol’s Application Modules, Credential Oracle Engine, and Credential API. 👨🏻‍💻 We are currently at the MVP stage → feel free to test our MVP here (get a wallet’s credit score calculation using Ethereum on-chain data), plus check our GitHub here and pitch deck here.Table of ContentIntroductionMarketProblemSolutionKey FeaturesValue for the EcosystemValue for CommunitySolution ExplainedThe AlgorithmTeamIntroductionNomis Protocol is an open-source protocol which provides dynamic APR, utilization rate and collateral size, which are calculated and balanced based on wallets’ on-chain data. The protocol operates on its own native token.Market$209BTotal value locked in DeFi as of Feb 11, 2022+338%TVL growth since last year$30BTVL in lending protocols as of Feb 11, 2022 Source: defillama.comProblemA lot of people still don’t have access to capital, especially in countries with underdeveloped economies. Also, people don’t want (or don’t understand why they need) to get a loan which is less than the collateral they need to use to back such a loan.SolutionWe aim to build a protocol which will use math and an AI-based prediction model to enable users with a positive on-chain credit score to take out crypto loans with less collateral. In the high-level concept, we want to give access to capital to more people around the world who can’t afford (or don’t want to go with) 110% collateral or more, but at the same time have a good credit score based on their on-chain history. Nomis Protocol will be an open-source protocol. Other teams can also build on top of it, and already existing high-TVL protocols can use our on-chain credit scores to balance their interest rates and collateral size individually for different wallets (users). 👨🏻‍💻 We are currently at the MVP stage → feel free to test our MVP here (get a wallet’s credit score calculation using Ethereum on-chain data), plus check our GitHub here and pitch deck hereKey Features🧮Fair Credit Score Fair and transparent calculated credit score based on on-chain data💎Optimized Collateral Reliable wallets can use less collateral against their loans🛠Open Source The protocol allows other developers build great products on top of it🌕$NUM The protocol features a native utility token that is used for governance and rewards💵$nUSD Borrowers take loans in stablecoins, backed with collateral and the protocol’s native token⚖️Protocol Governance The protocol is fully managed by DAO, and all future updates are made by votingValue for the Ecosystem🙋New Users A fair interest rate and reduced collateral will likely attract borrowers from other protocols and networks💰Network’s TVL Goes Wild More new users will bring more capital to the network, which will benefit all the network participants🏦Ecosystem Development Current ecosystem dApps can borrow capital for liquidity and development with better rate✨Ecosystem Growth If the protocol will finds its market fit, more devs will be interested in building new dApps and protocols on top of itValue for Community📈Mass Adoption Hopefully, crypto loans will become more accessible and affordable💫Access to Capital Users from underdeveloped economies will get access to affordable capital💻**Jobs for Developers ** More liquidity should attract more teams to build new products for the ecosystemSolution Explainedℹ️ The functional purpose of the solution is to calculate the reliability of a crypto borrower by analyzing the automatically collected digital credentials of the crypto wallet To predict a crypto borrower’s reliability on the basis of a wallet’s data and transaction history, we propose to use a trained artificial intelligence model, built on hierarchy analysis and pairwise comparisons methods. The choice of these methods is justified by insufficiency of data on defaults of crypto loans for construction of mathematical models of forecasting based on statistical approaches. The hierarchy analysis method seems to be more reasonable than the linear logic approach, as it solves multi-criteria problems in complex environments with hierarchical structures involving both tangible and intangible factors. Also, the method of hierarchy analysis is a closed logical construction, providing, with the help of simple rules, analysis of complex problems in all their diversity and leading to the best answer. Moreover, this method will make it possible to include all the knowledge we have on the problem in question in the dataset. This approach, from our point of view, is a balanced way of solving a quite difficult problem: forecasting when to make credit decisions under conditions of insufficient statistical data.The AlgorithmAt the first stage, we make a matrix of paired comparisons of parameters that affect the reliability of the crypto borrower in order to assess the importance of, among others, the parameters listed in the table below: The DAO members evaluate each pair of parameters against each other. Using a comparison technique, they must compare each parameter to the others and assign a value according to Saaty's pairwise comparison technique The use of DAO voting principles will make it possible to balance risk assessments using the paired comparison method. Implementating a DAO will create a revolution in the classical approach described by the mathematician Saaty in the construction of expert assessments using the pairwise comparisons method for making a balanced decision. This is because the experts are market participants themselves, and their interest in minimizing risks will be directly reflected in the mathematical model by voting when comparing parameters.Founding TeamFedor CherepanovMath & Blockchain DevPhD in Computer ScienceAuthor/co-author of more than 40 scientific publications on AIHolder of 8 patent certificates of artificial neural network softwaresDeveloped multiple blockchain solutions on Solidity (Etherium, Polygon), and Exonum (Rust) framework.Alex BarabashLinkedIn • TelegramMSс in Mathematics and EconomicsDeep AI and ML skills15 years of experience in neural network and expert systems researchCTO — 5 yearsFounder: GoRecruit, Vetlan, ModusArty ShatilovTwitter • LinkedIn • TelegramMSc in International Business DevelopmentProduct Manager — 3 yearsUI/UX designer — 4 years4+ years in blockchain and web3Founder of KREO — AR/VR studio; clients: Huawei, Yandex, Mazda, RedBullEx. co-founder, Product Designer and Product Manager at KickCity — web3 event/ticketing platform (Ethereum)Nickolay ChebotovBackend DeveloperDegree in CS.NET developerActive open source contributor5 times blockchain hackathons awardee

## Publication Information

- [unchase](https://paragraph.com/@unchase/): Publication homepage
- [All Posts](https://paragraph.com/@unchase/): More posts from this publication
- [RSS Feed](https://api.paragraph.com/blogs/rss/@unchase): Subscribe to updates