Dialog Engine Cancellation Score Adjustment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing spoken dialog systems often misinterpret user inputs, particularly when users attempt to cancel requests, leading to unnecessary further clarification or actions, as they fail to accurately classify cancellation commands.

Innovation Solution

A method that evaluates spoken terms for potential cancellation commands by assigning cancellation scores based on user actions following the input, allowing the system to learn from user interactions and adjust scores accordingly, enabling proper processing of cancellation inputs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the dialog system requests further clarification or refinement after user input, then the system can obtain more accurate information, but the user may need to cancel their initial request leading to misinterpretation of cancellation commands

Engineering Contradiction:

Improveaccuracy of input classificationVSAvoiduser experience

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system dynamically changes the parameter of cancellation score thresholds based on dialog state. When the system requests clarification, it adjusts the threshold for recognizing cancellation commands, allowing flexible adaptation to different dialog contexts and reducing misinterpretation of user inputs

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system uses feedback from user actions (such as opening applications or performing tasks) to update cancellation scores for spoken terms. This feedback mechanism allows the system to learn from user behavior and improve its ability to accurately distinguish cancellation commands from other inputs

Inventive Principle:

Principle #23Feedback

2Adaptability or versatility

If the system assigns fixed cancellation scores to spoken terms, then the classification process is simple, but the system cannot adapt to learn new potential cancellation terms

Engineering Contradiction:

Improveability to learn new cancellation termsVSAvoidcomplexity of score assignment process

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system performs self-learning by automatically updating cancellation scores based on observed user actions. When a user speaks a term followed by actions inconsistent with the original request (such as opening a different application), the system autonomously adjusts the cancellation score for that term without requiring manual programming

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The cancellation scores are made dynamic rather than static. The system continuously updates scores based on new user interactions and dialog contexts, allowing the classification process to adapt to changing user behavior patterns while maintaining a manageable complexity through structured score adjustment mechanisms

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS9767801B1Intelligently canceling user input

Publication Date: 2017.09.19 GOOGLE LLC

AI summary

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for intelligently cancelling user inputs. In one aspect, a requests input by a user is received by a dialog engine. A prompt or notification regarding the request is output by the dialog engine. That the user has taken an action in response to the prompt or notification is determined by the dialog engine. Based on the action taken by the user, that the response corresponds to a potential cancellation command is determined by the dialog system.