Virtual Assistant Missed Utterance Resolution System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intelligent virtual assistants often fail to respond to 'missed utterances' – commands or questions that are not recognized or cannot be answered due to limitations in their databases or understanding, leading to suboptimal user experience.
Innovation Solution
A system and method for identifying, categorizing, and prioritizing missed utterances by extracting features such as entities and key phrases, and providing them to a service for resolution, which updates the virtual assistant's instructions to enable better response capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the virtual assistant maintains a limited database and instruction set, then device complexity and resource usage are reduced, but the ability to respond to user commands deteriorates
Solution Approach 1:
The system performs preliminary action by collecting and logging missed utterances in advance, categorizing them by type and frequency, and prioritizing them before resolution. This allows the virtual assistant to systematically expand its capabilities based on actual usage patterns without immediately increasing complexity
Solution Approach 2:
The virtual assistant implements self-service through automated logging, categorization, and prioritization of missed utterances. The system automatically identifies gaps in its capabilities and generates resolution priorities without requiring manual analysis, enabling autonomous improvement while managing complexity
2Reliability
If the virtual assistant attempts to handle all possible commands, then service quality improves, but processing time and computational resources increase
Solution Approach 1:
The system applies partial action by focusing resolution efforts on the most frequent and important missed utterance categories. Rather than attempting to resolve all possible missed commands simultaneously, the system prioritizes high-impact areas based on logging data, improving service quality efficiently
Solution Approach 2:
The system implements feedback loops where missed utterances are logged, analyzed, and used to generate prioritized resolution lists. This continuous feedback mechanism allows the virtual assistant to iteratively improve service quality based on actual performance data and user interactions
Data Source
AI summary
An example computing device can include a processing resource and a memory resource storing instructions thereon, the instructions executable by the processing resource to: log commands directed to the computing device, identify missed utterances from the logged commands, extract features from the missed utterances, wherein the features include an entity and a key phrase associated with the missed utterances, generate a list of the missed utterances based on a priority associated with the features associated with each of the missed utterances, and provide a portion of the missed utterances to a service for resolution, wherein the portion of the missed utterances are above a threshold priority within the list of the missed utterances.