Dynamic Data Rating Engine for Content Supervision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current content filtering and monitoring systems are inadequate in rating the importance of online activities, particularly for older children, as they often block access to inappropriate data without providing supervisors with meaningful insights into the user's internet usage, leading to frustration and an overwhelming log of non-blocked data that fails to highlight important information.
Innovation Solution
A system that monitors and rates current instances of data based on preceding and succeeding instances, using a rating engine to assign importance scores and update ratings dynamically, while also considering characteristics like links, language, and time of access, and decaying importance over time if no new activity occurs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If content filtering systems block access to inappropriate data, then safety is improved, but supervisors lose meaningful insights into user internet usage
Solution Approach 1:
The system segments internet usage data into individual instances (pages, files, videos) and assigns independent ratings to each, allowing supervisors to see detailed usage patterns without being overwhelmed by blocked content. Each data instance is evaluated on its own merits while considering context from surrounding instances.
Solution Approach 2:
The rating system acts as an intermediary between the content filtering system and supervisors. Instead of simply blocking content or providing raw logs, the system intermediates by analyzing and rating each data instance, translating raw usage data into meaningful insights that help supervisors understand user behavior patterns.
2Loss of information
If the system logs all non-blocked data, then complete usage information is obtained, but the log becomes overwhelming and important information is lost
Solution Approach 1:
The system divides the complete usage log into individual data instances, each with its own rating. This segmentation allows supervisors to focus on rated instances rather than processing entire logs, reducing complexity while maintaining complete information availability.
Solution Approach 2:
Different data instances receive different ratings based on their local characteristics and context. Important or suspicious instances receive higher ratings and are highlighted, while routine instances receive lower ratings. This local differentiation allows supervisors to focus attention where needed without reviewing every log entry.
3Productivity
If the system rates data based on isolated instances, then processing speed is maintained, but the rating accuracy decreases
Solution Approach 1:
The system merges the evaluation of individual data instances with the analysis of surrounding context. By considering preceding and succeeding instances, the system combines isolated instance processing with contextual analysis, achieving both speed and accuracy through efficient sequential processing.
Solution Approach 2:
The system performs preliminary rating of individual instances quickly, then uses those ratings to inform the evaluation of surrounding instances. This preliminary action allows the system to maintain processing speed while improving accuracy through iterative refinement based on contextual information.
Data Source
AI summary
A computer-implemented method for rating a current instance of data is described. An activity occurring on a computing system is monitored. The activity includes the current instance of data. A rating is assigned to the current instance of data. A determination is made as to whether the activity includes at least one additional instance of data. When the activity includes at least one additional instance of data, the rating assigned to the current instance of data is updated. The updated rating is based on the content of the at least one additional instance of data.


