Context Stack for Multi-Turn Voice E-Commerce Conversations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current natural language processing systems for e-commerce interfaces lack contextual recall, resulting in robotic conversations and inability to reference prior utterances for identifying known information, and are limited to 'walled gardens' with specific software and hardware capabilities.
Innovation Solution
A system and method that generate a context stack with context entries including root intent elements, entity lists, and dialogue stacks to locate missing semantic elements, enabling intent flow execution requests and improving contextual understanding in multi-turn conversations across various platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If current natural language systems are used to provide responses to general queries, then the system can handle specific tasks, but the system provides very limited contextual recall and makes conversations sound robotic
Solution Approach 1:
The system performs preliminary actions by generating a context stack before processing user utterances. This context stack is built in advance to contain relevant contextual information from previous dialogues, enabling the system to recall and utilize this information when generating responses, thereby improving conversational naturalness while maintaining access to contextual data
Solution Approach 2:
The patent introduces a context stack as an intermediary data structure between the user utterance processing and the response generation components. This context stack serves as a mediator that stores and manages contextual information, allowing the system to bridge the gap between limited contextual recall capabilities and the need for natural, context-aware conversations
2Measurement precision
If the system reviews context entries to locate missing semantic elements, then the system can identify known information from prior utterances, but the processing complexity increases
Solution Approach 1:
The patent segments the context processing into distinct components: context entries are divided into root intent elements, entity lists, and dialogue stacks. This segmentation allows the system to review and locate missing semantic elements more efficiently by searching through organized, modular context entries rather than unstructured data, improving entity identification accuracy while managing processing complexity through structured organization
Solution Approach 2:
The system adds a new dimension to context processing by implementing a multi-layered context stack structure with root intent elements, entity lists, and dialogue stacks. This dimensional organization transforms the context processing from a flat search problem into a structured, hierarchical search, enabling more precise entity identification through systematic review of context entries across multiple organizational layers
Data Source
AI summary
Systems and methods for e-commerce systems using natural language understanding are described. A computing device is configured receive a user utterance including at least one identified semantic component and at least one missing semantic component and generate a context stack including a set of context entries. Each of the context entries includes a root intent element, an entity list element, and a dialogue stack and each context entry in the set of context entries is associated with one of a user utterance or a system utterance. The computing device is further configured to review at least one context entry in the set of context entries to locate the at least one missing semantic element within the dialogue stack and generate an intent flow execution request including the at least one semantic element from the first speech data and the missing semantic element.