Digital Assistant Availability Scoring for Context-Aware Interaction Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital assistants fail to account for varying user availability levels across different interaction topics and scenarios, leading to frustrating user experiences and potential abandonment of their use.
Innovation Solution
A system and method for predicting user availability by collecting real-time and historical data to determine optimal interaction times and execute plans with high compatibility using an I/O device, incorporating learning mechanisms to refine user availability patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If digital assistant initiates interactions without considering user availability, then interaction frequency increases, but user experience deteriorates and frustration increases
Solution Approach 1:
The system performs preliminary actions by collecting real-time data about user context (location, activity, time) and historical interaction data before initiating any interaction. This advance preparation allows the digital assistant to predict user availability and select appropriate interaction times, ensuring high-frequency interactions do not occur during unsuitable moments, thereby maintaining user experience quality
Solution Approach 2:
The system dynamically adjusts interaction timing based on real-time user availability assessment. Instead of using fixed interaction schedules, the digital assistant continuously monitors user context and adapts interaction initiation decisions dynamically, allowing interaction frequency to vary according to user state while maintaining experience quality
2Device complexity
If digital assistant uses single availability score for all topics, then system complexity is reduced, but accuracy of availability prediction deteriorates
Solution Approach 1:
The system segments the availability assessment by creating topic-specific availability scores instead of using a single unified score. Different availability scores are generated for different interaction topics based on user preferences and context, allowing the system to accurately predict user availability for each specific topic while maintaining manageable system complexity through modular score generation
3Productivity
If digital assistant interacts during all time slots, then coverage of interaction opportunities increases, but relevance of interactions decreases
Solution Approach 1:
The system changes the parameter of interaction timing by determining optimal time slots based on user availability patterns, preferences, and context. Instead of uniformly distributing interactions across all time slots, the digital assistant adjusts interaction timing parameters to align with user availability, ensuring both broad coverage of opportunities and high relevance to user state
Data Source
AI summary
A method for predicting availability of a user for interaction with a digital assistant, comprises: collecting real-time data about the user and historical user-agent interaction data when the user is present in proximity to the digital assistant as determined based on information collected by a sensor of I/O device executing the digital assistant; determining a plurality of current user availability scores based on the collected data to derive optimal times for initiation of interactions between digital assistant and the user; determining a user availability indicating the optimal time slots for executing a plan having a highest compatibility with the current user availability score from amongst a plurality of plans executable by the digital assistant via the I/O device; generating a proposed schedule for executing the plan based on the user availability pattern and executing a plan by the digital assistant via the I/O device based on the schedule.