Dialog Manager Segmentation for Conversational Understanding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for human-computer interaction face challenges in updating components to adapt to new tasks, conversing naturally with users, providing access to conversational understanding capabilities, and identifying irrelevant information when processing natural language inputs that request tasks.
Innovation Solution
A computer-implemented system with a dialog manager and task completion handler separates the functionality of determining intent and gathering information, allowing for third-party plugins and delayed decisions on user experience, enabling efficient handling of new intents and conversational styles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conversational understanding capabilities are combined in a single component, then the system can process natural language inputs, but it becomes difficult and time-consuming to update the component to adapt to new types of tasks
Solution Approach 1:
The patent segments the conversational understanding system into separate components: a dialog manager that handles conversation flow and a task completion handler that processes specific tasks. This segmentation allows each component to be updated independently, improving adaptability to new tasks without increasing overall system complexity.
Solution Approach 2:
The patent extracts the task completion functionality from the dialog manager into a separate task completion handler. This extraction enables the dialog manager to focus on conversation management while the task completion handler handles task-specific logic, making it easier to update task handlers for new tasks without affecting the core dialog management functionality.
2Adaptability or versatility
If all conversational understanding capabilities are integrated in one component, then the system can naturally converse with users, but it becomes difficult to provide third-party developers access to conversational understanding capabilities
Solution Approach 1:
The patent divides the system into a dialog manager and task completion handler, creating clear boundaries and interfaces. This segmentation allows third-party developers to access and extend specific components (such as adding new task handlers) without needing to understand or modify the entire system, thereby improving accessibility while maintaining manageable complexity.
Solution Approach 2:
The dialog manager is designed as a universal component that can work with multiple different task completion handlers. This multi-functionality allows third-party developers to create custom task handlers that integrate with the existing dialog manager, providing broad access to conversational understanding capabilities without increasing the core system's complexity.
3Reliability
If the system gathers all identified information, then it can fulfill user intents, but it may collect potentially irrelevant information that does not need to be gathered
Solution Approach 1:
The patent implements dynamic information gathering where the task completion handler determines which identified information to actually collect based on the current conversation context and task requirements. This dynamic approach ensures that only relevant information is gathered, maintaining reliability while avoiding collection of unnecessary information.
Solution Approach 2:
The patent applies local quality by allowing different parts of the information gathering process to have different characteristics. The dialog manager identifies potential information needs globally, while the task completion handler selectively gathers only the locally relevant information needed for specific tasks, filtering out irrelevant information.
Data Source
AI summary
Exemplary embodiments relate to methods, mediums, and systems for managing a conversation. In an embodiment, a computer-implemented input interface is provided to receive an input comprising information in natural language. A dialog manager is configured to determine an intent of the input, determine information to fulfill the intent, and identify one or both of information available to the dialog manager or information that is unavailable to the dialog manager. A conversational understanding document documents the intent and the identified information. An output interface forwards the conversational understanding document towards a task completion handler separate and distinct from the dialog manager. Other embodiments are described and claimed.