Knowledge Graph Path Search for Indirect Utterance Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing artificial intelligence applications for natural language-based interaction with end users is challenging, particularly in establishing logical connections between indirect user utterances and intended goals, as human-authored dialog flows are costly and often fail to anticipate user intentions.
Innovation Solution
A method that parses indirect user utterances into logical forms, maps them to subgraphs in a knowledge graph, searches for a best path to generate an intermediate logical form, and performs system actions if the graph structure matches a transaction threshold, enabling the conversational system to handle implicit requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If human-authored dialog flows are used to guide conversation, then the system can handle explicit user goals, but it fails to anticipate indirect user utterances and requires costly development time
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary structure between user utterances and dialog flows. The knowledge graph contains entities, attributes, and relationships that serve as a mediator to connect indirect user statements (e.g., 'I am retired') with relevant transactional goals (e.g., 'retiree discount'). This intermediary enables the system to infer user intentions without requiring explicit dialog flow rules for every possible utterance.
Solution Approach 2:
The system performs self-service by automatically inferring user goals from indirect utterances through knowledge graph traversal and path finding. Instead of requiring pre-programmed responses for every possible user input, the system autonomously analyzes the utterance, searches the knowledge graph for relevant connections, and determines appropriate transactions or information to provide, reducing the need for extensive human-authored dialog flows.
2Adaptability or versatility
If hand-crafted dialog flows are developed with subject matter expertise, then coverage of explicit goals improves, but development time and cost increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-structuring domain knowledge into a knowledge graph with entities, attributes, and relationships before the conversational system needs to handle user utterances. This preliminary organization of information (e.g., linking 'retired' status to 'retiree discount' benefits) enables rapid inference during conversation without requiring time-consuming dialog flow development for each scenario.
Solution Approach 2:
The knowledge graph serves as a universal structure that can handle multiple types of user utterances and goals through a single framework. Rather than creating separate dialog flows for different scenarios, the same knowledge graph infrastructure supports various transactions and information queries by traversing different paths based on the user's input, making the system versatile without proportionally increasing development time.
3Adaptability or versatility
If the system searches for best paths in knowledge graph to connect indirect utterances to goals, then handling of implicit requests improves, but processing complexity increases
Solution Approach 1:
The patent segments the complex task of understanding indirect utterances into distinct processing steps: parsing the utterance to extract entities and attributes, mapping these to knowledge graph nodes, searching for paths between nodes, and generating responses based on found paths. This segmentation of the processing pipeline makes the overall complex task more manageable and implementable through modular components.
Data Source
AI summary
A method, apparatus and computer program product for establishing a logical connection between an indirect utterance and a transaction is described. An indirect utterance is received from a user as an input to a conversational system. The indirect utterance is parsed to a first logical form. A first set of predicates and terms is mapped from the first logical form to a first subgraph in a knowledge graph. A second set of predicates and terms is mapped from a second logical form belonging to a transaction to a second subgraph of the knowledge graph. A best path in the knowledge graph between the first subgraph and the second subgraph is searched for while transforming the first logical form using the node and edge labels along the best path to generate an intermediate logical form. A system action is performed for a transaction if a graph structure of the intermediate logical form matches the graph structure of the logical form of the transaction above a threshold.