Dialog System Semantic Mapping for Rich Intent Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current dialogue systems are limited in handling richer semantic human communication, requiring step-by-step API requests and being unable to efficiently manage multiple API calls, leading to an inconvenient user experience.
Innovation Solution
A dialog system utilizing a recurrent neural network to extract semantics from user requests and organize them into a resource graph, allowing for the optimization and execution of multiple API calls as a single response action, trained directly on API sequences rather than human-defined semantics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the system uses conventional step-by-step API handling, then the system structure is simple, but the user experience deteriorates due to inability to handle richer semantic communication
Solution Approach 1:
The patent introduces an intermediary component that translates human semantics into machine service APIs. This intermediary layer enables the system to understand richer semantic communication without requiring complete restructuring of the underlying system architecture, thus resolving the contradiction between adaptability and complexity.
Solution Approach 2:
The system segments the complex task of handling multiple APIs into manageable components: semantics extraction, service composition, and API execution. This segmentation allows the system to handle richer semantics while maintaining a structured and manageable system architecture.
2Adaptability or versatility
If the system lists all valid API combinations exhaustively, then complete service coverage is achieved, but the system complexity becomes enormous and impractical
Solution Approach 1:
The system performs preliminary actions by pre-defining service compositions and API relationships in a knowledge base. When a user request arrives, the system queries this pre-prepared knowledge base rather than exhaustively analyzing all possible API combinations in real-time, thus achieving complete service coverage without enormous complexity.
Solution Approach 2:
Instead of managing all possible API combinations directly, the system creates simplified representations (copies) of service compositions in the knowledge base. These copies capture the essential relationships and can be quickly queried and executed without dealing with the full complexity of all possible combinations.
3Measurement precision
If the system asks for information step-by-step, then information accuracy is ensured, but communication convenience deteriorates significantly
Solution Approach 1:
The system performs preliminary semantics extraction and analysis before executing API calls. By understanding the user's intent and required information upfront through semantic parsing, the system can retrieve accurate information in a single operation rather than asking step-by-step, thus maintaining information accuracy while improving communication convenience.
Data Source
AI summary
A dialog system and intelligent personal assistant capable of semantic-understanding mapping between user intents and machine services includes an interface to receive a voice or text request from a user. Semantics of the request including at least one of entities, intent, or context are extracted from the user's request. A sequence of action features is selected based on the extracted semantics. A sequence of application programming interfaces (APIs) corresponding to the sequence of action features is then executed to generate a result. An action sequence optimizer may optimize the sequence of action features based on user configuration. The examples provide a technical solution to model a richer semantic-understanding mapping between user intents and available APIs, that will be greatly improving user experience in the spoken dialogue system, as core of personal assistant.