Knowledge Anchored AI Tree for Explainable Ecommerce Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AI systems require extensive labeled data, struggle to explain their decision-making processes, and lack basic common sense knowledge, making them inefficient for handling ecommerce transactions across multiple merchant websites without human intervention.
Innovation Solution
A knowledge-anchored artificial intelligence (KAAI) system that uses a tree-based structure to process queries, learns semantic graphs from web applications automatically, and integrates with user preferences, allowing for zero-dependency onboarding and handling ecommerce transactions through a digital assistant.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional AI systems are used for ecommerce transactions, then they can process data and make decisions, but they require extensive labeled data and cannot explain their decision-making processes
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary layer between the AI system and the transaction processing. This knowledge graph stores structured domain knowledge about ecommerce transactions, products, and services, allowing the AI to make decisions based on explicit knowledge rather than opaque neural network predictions. The knowledge graph serves as a mediator that provides explainable reasoning paths for automated decisions.
Solution Approach 2:
The patent segments the AI system into multiple components: a neural network for pattern recognition, a knowledge graph for structured knowledge storage, and a reasoning engine for decision-making. This segmentation allows each component to specialize in specific tasks while maintaining transparency in the overall decision-making process, resolving the contradiction between automation and explainability.
2Adaptability or versatility
If traditional AI systems are deployed across multiple merchant websites, then they can handle diverse ecommerce transactions, but they lack basic common sense knowledge and require extensive training data
Solution Approach 1:
The patent pre-populates the knowledge graph with common sense knowledge about ecommerce transactions, product categories, and merchant operations before deployment. This preliminary action allows the AI system to immediately handle diverse transactions across multiple merchants without requiring extensive labeled training data for each specific scenario, as the foundational knowledge is already in place.
Solution Approach 2:
The knowledge graph is designed as a universal structure that can accommodate multiple merchant websites and diverse transaction types through a common schema. This universal knowledge representation allows the AI system to adapt to different merchants and transaction scenarios without requiring separate training data for each, resolving the contradiction between versatility and data quantity.
3Adaptability or versatility
If users need to onboard to multiple merchant websites for ecommerce purchases, then they can access various products and services, but they must manage multiple passwords which leads to frustration and inefficiency
Solution Approach 1:
The patent merges the authentication and user management functions across multiple merchant websites into a unified system. By integrating user profiles, preferences, and authentication credentials into a centralized knowledge graph, the system allows users to access multiple merchants through a single onboarding process, eliminating the need to manage multiple passwords while maintaining secure access to diverse ecommerce platforms.
Data Source
AI summary
For one embodiment of the present disclosure, a knowledge anchored artificial intelligence (KAAI) system and method are disclosed herein. A computer implemented method includes receiving a query to KAAI system; creating the necessary actions to respond to the query using a KAAI tree (literally by traversing the KAAI tree) to create the chain of thoughts and also make sure that the actions are anchored using existing knowledge graph in the system;maintaining a KV memory called session which will be preserved and changed while processing the query and traversing the tree; At each node in the tree execute Compute Agent logic i. Get data from the backend ii. Call into API iii. Run data transformations; b. May further interact with the user through UI; c. May amend the tree if needed; d. Update the context memory (session) e. Update the values in the tree.