Human-Machine Conversation Knowledge Graphs for Fewer Interaction Turns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech interaction in smart home devices experiences poor multi-dimensional conversation understanding and intelligent response, leading to a suboptimal interaction experience due to the lack of a common cognitive context between the machine and the user.
Innovation Solution
A human-machine multi-turn conversation method involving the establishment of a knowledge graph to determine support degrees of child nodes based on historical query records, with semantic information output based on threshold values, allowing for intelligent and efficient interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple turns of conversation are used to identify user intention, then the machine can understand user intent, but the interaction experience deteriorates due to excessive conversation turns
Solution Approach 1:
The system performs preliminary actions by establishing a knowledge graph of user conversation behavior information before actual interaction occurs. This pre-built structure enables the machine to predict and understand user intentions without requiring multiple conversational turns, thus resolving the contradiction between understanding accuracy and interaction efficiency
2Extent of automation
If a knowledge graph of user conversation behavior is established, then the machine can understand user intentions directly, but the device complexity increases
Solution Approach 1:
The knowledge graph is segmented into hierarchical structures with nodes representing different levels of conversation behavior information. This segmentation allows the complex user behavior data to be organized into manageable, structured components that can be efficiently processed while maintaining high automation capability
Solution Approach 2:
The knowledge graph serves as an intermediary structure between raw user input and the machine's understanding system. By introducing this intermediate representation layer, the system can handle complex user intentions without requiring equally complex processing mechanisms, thus balancing automation capability with device complexity
3Manufacturing precision
If support degree calculation based on historical query records is performed, then the response relevance improves, but the processing time increases
Solution Approach 1:
The system performs preliminary action by pre-calculating and storing support degrees for various node relationships in the knowledge graph based on historical query records. This pre-computation enables rapid retrieval during actual interaction, improving response relevance without incurring processing time delays during conversation
Data Source
AI summary
The present disclosure relates to a human-machine multi-turn conversation method and system for human-machine interaction, and an intelligent apparatus. The method includes: S1, establishing a knowledge graph of user conversation behavior information; S2, determining, according to information currently input by a user, a node corresponding to the information currently input and at least one child node of the node in the knowledge graph; S3, calculating a support degree of the at least one child node relative to the node according to the number of times of querying the node and the number of times of querying both the at least one child node and the node in a historical query record of the knowledge graph; and S4, determining whether to output semantic information of the at least one child node by determining a size relation between the support degree and a preset support degree threshold value.