# How AI and Blockchain Team Up to Predict What You'll Buy > This patent describes a system where different companies can privately share insights from their customer data using AI and blockchain to build a better model for predicting if a customer will buy a product. - **Patent:** US 20230419182 - **Original title:** Methods and systems for imrpoving a product conversion rate based on federated learning and blockchain - **Owner:** Hangzhou Tonghuashun Data Processing Co - **Status:** Active - **Times cited:** 12 - **Field:** software, ai_ml, ecommerce, telecommunications, finance ## What it does This system helps improve how well businesses can predict if a customer will buy a product. A central 'supervisor node' receives a request from an 'initiator node' (like an online store) that wants to build a better prediction model (Claim 1). The supervisor then broadcasts this request across a secure 'blockchain federation' to find other 'participant nodes' (other businesses) willing to contribute their data insights. Instead of sharing raw customer data, these nodes only share 'representation data' – a summary or pattern of their customer information. Based on this shared representation data, the supervisor decides the best way to combine their knowledge, choosing between a 'longitudinal' or 'horizontal' federated learning strategy (Claim 3). Finally, the supervisor coordinates the initiator and participant nodes to collaboratively train a 'conversion rate model' without any single party seeing all the raw data. For example, an e-commerce site could use this to predict if a user will purchase a specific item after viewing it, by combining its own data insights with those from other related businesses. ## What it does NOT cover - Does not cover traditional machine learning systems where all raw user data is collected and processed in one central location. - Does not cover federated learning systems that do not use a blockchain for coordination, broadcasting requests, or recording rewards. - Does not cover predicting general user behaviors; it specifically focuses on predicting a 'product conversion rate' for a 'preset product'. - Does not cover systems that directly share raw user data between nodes; it emphasizes sharing 'representation data' instead. - Does not cover a system without a designated 'supervisor node' to coordinate the federated learning process. ## The clever bit The clever part is the integration of blockchain for secure and transparent coordination of federated learning, especially the dynamic selection of a 'longitudinal' or 'horizontal' learning strategy based on how similar the participating datasets are. ## Real-world examples 1. Personalized product recommendations on e-commerce platforms 2. Targeted advertising campaigns in online retail 3. Fraud detection systems in financial services 4. Customer churn prediction in subscription services ## Why it matters Predicting what customers will buy is crucial for businesses, but privacy concerns often limit how much data can be shared and used. This patent addresses that challenge by combining federated learning, which allows AI models to learn from decentralized data without direct sharing, with blockchain technology for secure and transparent coordination. This approach could enable more accurate prediction models by leveraging a wider range of data insights, while still protecting user privacy and complying with data regulations. ## Frequently asked questions ### What does How AI and Blockchain Team Up to Predict What You'll Buy cover? This patent describes a system where different companies can privately share insights from their customer data using AI and blockchain to build a better model for predicting if a customer will buy a product. ### Who owns patent US 20230419182? This patent is owned by Hangzhou Tonghuashun Data Processing Co. ### When does this patent expire? This patent is expected to expire on May 22, 2043, when the invention enters the public domain. ### What is patent US 20230419182 cited by? This patent has been cited by 12 later patents that build on its ideas. ### What problem does this patent solve? Predicting what customers will buy is crucial for businesses, but privacy concerns often limit how much data can be shared and used. This patent addresses that challenge by combining federated learning, which allows AI models to learn from decentralized data without direct sharing, with blockchain technology for secure and transparent coordination. This approach could enable more accurate prediction models by leveraging a wider range of data insights, while still protecting user privacy and complying with data regulations. ### What does this patent NOT cover? Does not cover traditional machine learning systems where all raw user data is collected and processed in one central location. **Full plain-English explainer:** https://patentbrief.org/patent/us/20230419182/methods-and-systems-for-imrpoving-a-product-conversion-rate-based-on-federated-l **Original patent:** https://patents.google.com/patent/US20230419182 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How Training Systems Share AI Knowledge Without Sharing Private Student Data](https://patentbrief.org/patent/us/11915111/federated-machine-learning-in-adaptive-training-systems) — This patent describes a system where different student training centers can improve their AI models by sharing general learning patterns, not private student data, to make training better for everyone. - [Sharing AI Knowledge Between Private Datasets Using Synthetic Data](https://patentbrief.org/patent/us/12039012/systems-and-methods-for-heterogeneous-federated-transfer-learning) — This patent describes a method for two separate AI systems to share learned information by generating artificial data based on common features, without directly exchanging their private training data. - [Training AI on Private Data Without Seeing It](https://patentbrief.org/patent/us/12518214/distributed-machine-learning-systems-including-generation-of-synthetic-data) — This patent describes a way to train artificial intelligence models using private data stored on many separate computers, by generating fake data that mimics the real data's patterns, so the private data itself never leaves its original location. - [Training AI Models Together with Unlabeled Data Using a Teacher](https://patentbrief.org/patent/us/20220012637/federated-teacher-student-machine-learning) — This patent describes a way for multiple AI systems to learn together from data that hasn't been manually labeled, using a 'teacher' AI to create temporary labels for a 'student' AI. - [How Devices Train Shared AI Models While Keeping Your Data Private](https://patentbrief.org/patent/us/12443890/partially-local-federated-learning) — This patent describes a method for training a machine learning model across many devices, where each device keeps some parts of the model and its data private, only sharing updates for the common, global parts of the model.