Pop-up Content Prediction Using Time-Weighted Usage Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic apparatuses struggle to accurately predict user usage patterns due to equal weighting of recent and old usage data without considering time, leading to inaccurate reflections of recent usage history.
Innovation Solution
Applying different weighted values to user's usage state information based on time sections, with higher weights for recent data to improve prediction accuracy, and grouping usage state information to reflect changing user interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If equal weighted values are applied to all usage data regardless of time, then the calculation process is simple, but the prediction accuracy of user behavior deteriorates
Solution Approach 1:
The patent applies different weighted values to usage data based on time parameters, where more recent usage data receives higher weights and older data receives lower weights. This transforms the uniform weighting parameter into a time-dependent parameter, improving prediction accuracy while maintaining reasonable computational complexity through the formula: predicted usage count = Σ(actual usage count × weighted value)
Solution Approach 2:
The weighting system dynamically adjusts the importance of different usage data points based on their temporal characteristics. Rather than using static equal weights, the system makes weights dynamic by assigning higher values to recent interactions and lower values to historical interactions, allowing the prediction model to adapt to changing user behavior patterns
2Measurement precision
If different weighted values are applied based on time to improve prediction accuracy, then the reflection of recent usage history improves, but the complexity of data processing increases
Solution Approach 1:
The patent introduces time as a parameter that modifies the weight assigned to each usage data point. By changing from uniform weights to time-based weights, the system achieves better reflection of recent usage patterns. The weighted value is determined by the time elapsed since each usage event, creating a gradient where recent events have higher influence
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An electronic apparatus is provided. The electronic apparatus includes a memory configured to store usage state information on a pop-up content, and a processor configured to divide a time section based on interaction information of a user regarding the electronic apparatus, group the usage state information stored in the memory by each time section, predict or identify a usage pattern regarding a pop-up content by applying a different weighted value to each group, and identify whether to provide the pop-up content based on the predicted or identified usage pattern.