🗄️InRewards Network

InRecommendations — AI-driven decentralized recommendations system.

InRecommendations is the world’s first decentralized crowdsourced recommendation system, which is designed to suit the needs of e-commerce and uses the blockchain for quality control and security of the data.

How It Works - for Merchants:

• Merchants send the encoded data about the activities of their users to the InRecommendations system, and it sends them access to a cutting-edge recommendation system - free of charge.

• InRecommendations system offers product recommendations (“you may also like”) and rewards recommendations - with the possible ways of showing users the appreciation for their actions via a thoughtful reward: cryptocurrency, a cup of coffee, or a ticket to an event.

• InRecommendations learns on the go using data from thousands of merchants and processing millions of data points per day, which ensures the top quality of recommendations at all times.

• All the data received from merchants is stored on the decentralized file system IPFS. The metadata, including the ratings of data quality, is stored on the Ethereum blockchain. This type of structure makes the system eternally operable and available to the clients. If InRecommendations somehow ceases to work, with the support of InRewards DAO, a new system can always be created to use that data.

• If a merchant tries to cheat and sends fake data, it will be detected, and this information will go to the Ethereum blockchain.

• Merchants can see the previous user activity of every user - that’s why INVRecommendation provides the scoring of the user's reputation, which allows the merchants to offer a loan or a post-payment for the purchases.

InRecommendations — the Eyes and Brains of Online Stores

The recommendation system is a necessary component of the loyalty systems of the merchants - it enables them to incentivize higher order value and repeat purchases thus improving the ROI. An efficient recommendation system must contain huge amounts of data. This poses a problem for small and medium-sized businesses because they can’t acquire the necessary amount of traffic and sales to collect that data.

DMP Data Is Not Enough

Data management platforms, which aggregate data from various sources, are designed to provide merchants with the profiles of their users - social graph, interests, disposable income, etc. This information is useful but insufficient to make high-quality forecasts in e-commerce because user habits and consumption patterns are hard to figure out yet they contain data, which is crucial for decision-making. For instance, it’s impossible to predict what color of the device customers will prefer basing their assumptions on the data about their age and attraction to a certain car brand. InRecommendations is a decentralized DMP designed to collect, store and process the retail data using deep learning algorithms.

The control of the quality and security of the data is provided for by the blockchain. Every merchant is a source of huge amounts of data about their shoppers. 90% of this data remains unused. The best-case scenario is that the data gets sent to Google Analytics where a marketing specialist will go through its aggregated copy. InRewards will give merchants a solution, which will easily enable them to have all the user data at their disposal, including:

• general data types (viewed items, added to order items, order amount);

• industry-specific data types about the user interests (colors of garments, gadget screen sizes, food calories, etc.). This data will be collected and stored by the merchant on their servers. Merchants will receive a turnkey solution, which can be deployed on any server:

• every 24 hours the data is encoded and sent into IPFS;

• the merchant sends the IPFS-address containing the specific data and metadata to the InRecommendations smart-contract (number of the unique per period, merchant ID);

• the merchant can send this data to the smart-contract of any other organization, to which InRewards DAO has entrusted the private keys — it is necessary to avoid the dependence on the centralized InRecommendations software. As soon as the data is received in the smart-contract, InRecommendations runs several processes:

• checks the reputation of the merchant. If the reputation is unreliable, the weight of the data is set to “low”, or it is simply ignored;

• downloads the data from IPFS to the InRecommendations servers;

• starts the lock-up period (2 weeks at the beginning of the operations) and processes the received data via a moderation subsystem. The main purpose of moderation is to determine whether or not the data is accurate and came from reliable sources. Another purpose is to attach the appropriate weights to the data. To solve this problem, deep learning and AI are applied;

• after the lock-up period, InRewards credits the merchant with InRewards tokens for the valuable data and reduces the rating and repayments in case of fake data;

• the accurate data is used for deep learning algorithms.