Dialog System Feedback Classification for Answer Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dialog systems lack the ability to effectively classify user feedback and adjust confidence scores for question-answer pairs based on user input, leading to suboptimal performance in providing accurate and satisfactory responses.
Innovation Solution
A method is implemented where a dialog system evaluates user input to determine if it is feedback to a previously provided answer, classifies it as positive or negative, and adjusts the confidence score accordingly, providing alternative answers when necessary based on the feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the dialog system provides a fixed answer without feedback classification, then the system complexity is low, but the answer accuracy and user satisfaction deteriorate
Solution Approach 1:
The patent implements feedback classification by analyzing user inputs after providing answers. The system classifies feedback as positive or negative and adjusts confidence scores accordingly, enabling continuous improvement of answer accuracy through user responses while maintaining manageable system complexity through structured classification rules
Solution Approach 2:
The system dynamically adjusts confidence scores based on classified feedback. When negative feedback is detected, the system lowers confidence scores and provides alternative answers. This dynamic adaptation allows the system to improve answer accuracy over time without requiring complete system redesign
2Reliability
If the system classifies feedback and adjusts confidence scores, then the user satisfaction improves, but the processing time increases
Solution Approach 1:
The system performs partial feedback classification by focusing on key indicators of user satisfaction rather than analyzing every aspect of user input. This selective approach maintains user satisfaction improvement while reducing processing time compared to comprehensive analysis
3Measurement precision
If the system provides alternative answers based on negative feedback, then the answer quality improves, but the device complexity increases
Solution Approach 1:
The system pre-generates multiple candidate answers and stores them with associated confidence scores. When negative feedback is received, the system simply selects the next best alternative from the pre-prepared list rather than generating new answers, reducing computational complexity while maintaining answer quality
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a dialog system using user feedback. According to one implementation, a method includes receiving, by a dialog engine and from a computing device, a voice input; determining, by the dialog engine, that the received voice input is classified as feedback to an answer of a question; identifying, by the dialog engine, a predetermined feedback score associated with the feedback; and adjusting a confidence score associated with the question and the answer based on the predetermined feedback score.