Dynamic Dialog Control via Intent Probability and Confidence Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional automated response systems in customer contact centers, such as IVR systems, are inflexible and require customers to navigate lengthy menus, failing to understand unstructured user inputs and often necessitate painstaking detail in prompt creation and response handling.
Innovation Solution
A system and method that uses a domain model to automatically extract user intent from natural language inputs, dynamically traversing dialog flows and skipping unnecessary steps based on confidence levels, allowing for more robust natural language understanding and reduced customer effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional IVR systems are used to automate customer service, then employee overhead is reduced, but customer effort and interaction time increase due to lengthy menus
Solution Approach 1:
The patent replaces the traditional menu-navigation mechanical system with a natural language processing system that uses machine learning models to understand and respond to customer intent, eliminating the need for step-by-step menu traversal
Solution Approach 2:
The system dynamically changes the parameter of response handling from fixed menu options to flexible natural language interpretations, allowing the dialog system to adapt to various customer expressions and skip unnecessary interaction steps
2Device complexity
If traditional IVR systems are used with predetermined communication paths, then system control is simplified, but system flexibility and adaptability decrease
Solution Approach 1:
The patent transforms the static predetermined communication paths into dynamic dialog flows that adapt in real-time based on customer responses and confidence levels, allowing the system to navigate to appropriate handlers flexibly while maintaining controlled structure
Solution Approach 2:
The patent introduces a dialog control system with confidence level assessment as an intermediary between the natural language input and the predetermined handlers, enabling flexible routing while maintaining system control through structured decision-making
3Reliability
If traditional IVR systems require exact match responses, then system reliability is maintained, but system robustness and ability to handle unstructured inputs decrease
Solution Approach 1:
The patent replaces the exact-matching mechanical response verification system with a probabilistic natural language understanding system that assesses confidence levels and handles unstructured inputs through machine learning models
Solution Approach 2:
The system changes the parameter of response matching from binary exact-match to continuous confidence level assessment, enabling the system to maintain reliability through threshold-based decision-making while handling diverse natural language inputs
Data Source
AI summary
A system and method for engaging in an automated dialog with a user. A processor retrieves a preset dialog flow that includes various blocks directing the dialog with the user. The processor provides a prompt to the user based on a current block of the dialog flow, receives an action from the user in response to the prompt, and retrieves a classification/decision tree corresponding to the dialog flow. The classification tree has a plurality of nodes mapped to the blocks of the dialog flow. Each of the nodes represents a user intent. The processor computes a probability for each of the nodes based on the action from the user. A particular one of the nodes is then selected based on the computed probabilities. A target block of the dialog flow is further identified based on the selected node, and a response is output in response to the identified target block.