Virtual Agent Intent Disambiguation via Runtime Customization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Virtual agents are limited by pre-defined tasks and services, restricting user interaction to only those defined by the administrator, lacking the ability to autonomously adapt or understand user intents in real-time.
Innovation Solution
A virtual agent that analyzes user information, such as privacy rights, user roles, and action history, to dynamically define custom intents at runtime, using machine learning processes to interpret user commands and execute relevant actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If tasks and services are pre-defined by an administrator, then the virtual agent can provide structured and manageable services, but the user is restricted to only those pre-defined tasks and cannot interact beyond them
Solution Approach 1:
The patent applies dynamics by transforming static pre-defined intents into dynamic custom intents that are created at runtime. The system evolves from a fixed structure where tasks are predetermined to a flexible structure where intents are generated adaptively based on user input, application information, and metadata, allowing the virtual agent to respond to unforeseen user needs while maintaining system manageability
Solution Approach 2:
The patent implements self-service by enabling the virtual agent to autonomously create custom intents without requiring administrator intervention. The system uses machine learning models to automatically analyze user input, retrieve relevant application information and metadata, and generate appropriate intents and actions independently, freeing the system from constant administrative configuration while maintaining structured service delivery
2Adaptability or versatility
If the virtual agent uses pre-defined tasks, then the system structure remains simple and manageable, but the user experience lacks personalization and real-time adaptation
Solution Approach 1:
The patent replaces the mechanical system of manual administrator-defined task configuration with an automated machine learning-based intent generation system. Instead of relying on pre-programmed tasks, the system uses ML models to automatically analyze user input, retrieve contextual information from applications and metadata, and generate appropriate intents and actions dynamically, achieving real-time adaptation through automation
Solution Approach 2:
The patent implements feedback by using machine learning models that continuously learn from user interactions and system responses. The system analyzes user input, generates intents, executes actions, and uses the outcomes to refine future intent generation, creating a closed-loop system that adapts to user needs over time while maintaining autonomous operation
3Ease of operation
If custom intents are defined at runtime using machine learning, then user interaction becomes flexible and personalized, but the system complexity increases significantly
Solution Approach 1:
The patent applies segmentation by dividing the complex intent generation process into distinct modular components: user input processing, application information retrieval, metadata analysis, machine learning-based intent generation, and action execution. Each component handles a specific aspect of the process independently, making the overall complex system more manageable and easier to operate while maintaining high ease of use for users
Data Source
AI summary
The present disclosure is directed techniques for executing a task or service using a virtual agent. A method includes: executing, using a virtual agent, one or more tiers of a plurality of tiers of machine learning analysis to identify a desired action to be performed based on a user command, the user command being received from an external computing device; responsive to the one or more tiers of the plurality of tiers of machine learning analysis identifying a plurality of actions associated with the user command, determining a series of inquiries to present via the external computing device, wherein each inquiry of the series of inquiries is selected based on a number of actions associated with each inquiry, and wherein each subsequent inquiry in the series of inquires is based on a user response to a preceding inquiry; identifying, based on responses to the series of inquiries, the desired action to be performed; and executing the desired action to be performed.