Intelligent Virtual Assistant Selection via Collaborative Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in selecting the most appropriate speech-to-text or voice assistant service for their queries, leading to potential poor responses and an unpleasant user experience.
Innovation Solution
The system employs an intelligent virtual assistant selection (IVAS) service that receives user queries, determines the domain or task, identifies similar queries, and selects the best virtual assistant based on customer feedback and collaborative filtering to provide a response.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple virtual assistant services are made available to users, then service coverage and functionality are improved, but user confusion and difficulty in selection increase
Solution Approach 1:
The patent introduces an intermediary system that sits between the user and multiple virtual assistant services. This intermediary automatically analyzes user queries, identifies the most appropriate virtual assistant based on various criteria (query type, user preferences, service capabilities), and routes the query to the selected service. This mediator eliminates the need for users to manually select among multiple services, thereby maintaining service coverage while removing selection difficulty.
Solution Approach 2:
The system enables virtual assistant services to self-select by automatically analyzing incoming queries and determining which service is best suited to handle each query. The system uses machine learning models and collaborative filtering to allow services to compete and select themselves based on their strengths and past performance, eliminating manual user intervention in the selection process.
2Measurement precision
If users manually select virtual assistants for different queries, then control and precision are improved, but time consumption and complexity increase
Solution Approach 1:
The system performs preliminary actions by pre-configuring virtual assistant services with specific expertise areas, capabilities, and performance metrics. User preferences and historical data are pre-processed to create selection models. When a query arrives, the system quickly matches it against pre-established criteria rather than requiring real-time user analysis or manual selection, thereby maintaining precision while minimizing time loss.
Solution Approach 2:
The system implements feedback mechanisms where user selections, query outcomes, and satisfaction metrics are continuously collected and used to refine the automated selection process. This feedback loop allows the system to learn from past interactions and improve its selection precision over time, reducing the need for users to manually adjust selections while maintaining or improving accuracy.
3Reliability
If automated selection based on collaborative filtering is implemented, then selection accuracy is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The patent segments the automated selection system into distinct functional modules: query analysis module, collaborative filtering module, selection module, and feedback module. Each module handles a specific aspect of the selection process independently. This segmentation allows the system to achieve high selection accuracy through specialized processing in each module while managing overall complexity through modular architecture, making the system more maintainable and scalable.
Data Source
AI summary
A query is received from a user device. A domain and/or task is determined corresponding to the query. A set of similar queries to the query are identified using a collaborative selector. A plurality of virtual assistants (VAs) are ranked based on an average of customer feedback received from execution of the similar queries, the customer feedback including ratings of responses to the similar queries. One of the plurality of VAs is selected as being the one having a highest average of the customer feedback to use to respond to the query. The query is replied to using a selected response generated by the one of the plurality of VAs.