Digital Assistant Crowd-Sourcing Response Fallback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital assistants often fail to provide satisfactory responses to user requests due to limitations in natural language processing, knowledge base, and artificial intelligence, leading to user dissatisfaction and loss of confidence in the system.
Innovation Solution
The implementation of a method that crowdsources information from external sources, such as expert services and public forums, to generate responses to user requests when the digital assistant's real-time response mechanisms fail, utilizing a crowd-sourced knowledge base for future improvements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the digital assistant uses its own internal knowledge base and processing capabilities to respond to user requests, then the response time is fast and the system operates autonomously, but the accuracy and satisfaction of responses deteriorate when the internal capabilities are insufficient
Solution Approach 1:
The patent merges the digital assistant's internal knowledge base with external crowd-sourced information sources. When the internal knowledge base fails to provide a satisfactory response, the system queries external sources such as expert forums and information services to supplement or replace the internal response, thereby improving overall response satisfaction without requiring the assistant to have all knowledge internally stored.
Solution Approach 2:
The patent introduces a failure detection module as an intermediary that monitors the quality of responses generated by the digital assistant. When unsatisfactory responses are detected, this intermediary triggers a crowd-sourcing mechanism to retrieve additional information from external sources, acting as a mediator between the assistant's internal capabilities and external knowledge sources.
2Reliability
If the digital assistant queries external crowd-sourced information sources to improve response accuracy, then the knowledge base coverage improves, but the response time and system latency increase
Solution Approach 1:
The patent implements preliminary action by pre-configuring the digital assistant with a comprehensive internal knowledge base covering common queries and scenarios. This allows the system to respond quickly to frequent questions without needing to query external sources, and only initiates crowd-sourcing when the internal knowledge is insufficient, thereby minimizing time loss while maintaining high accuracy for complex or niche queries.
Solution Approach 2:
The patent employs feedback mechanisms where users can indicate whether a response was satisfactory. This feedback is used to train and improve the internal knowledge base over time, reducing the frequency of external queries needed. The system learns from user interactions to better predict when internal knowledge suffices and when external crowd-sourcing is necessary, optimizing the balance between response time and accuracy.
3Reliability
If the digital assistant implements sophisticated natural language processing and artificial intelligence to improve response quality, then the response accuracy improves, but the computational resources and processing complexity increase
Solution Approach 1:
The patent applies partial action by implementing a tiered response strategy. The digital assistant first attempts to answer using efficient internal knowledge base queries with minimal computational resources. Only when this partial approach fails to produce satisfactory responses does the system activate more computationally intensive crowd-sourcing mechanisms, thereby reducing overall energy consumption while maintaining high response quality for difficult queries.
Data Source
AI summary
A user request is received from a mobile client device, where the user request includes at least a speech input and seeks an informational answer or performance of a task. A failure to provide a satisfactory response to the user request is detected. In response to detection of the failure, information relevant to the user request is crowd-sourced by querying one or more crowd sourcing information sources. One or more answers are received from the crowd sourcing information sources, and the response to the user request is generated based on at least one of the one or more answers received from the one or more crowd sourcing information sources.