Zombie Account Detection via Subscription Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inactive online accounts, applications, and devices ('zombie accounts' and 'zombie devices') pose a burden on system infrastructure and privacy risks due to continued resource usage and potential unauthorized access, as users stop accessing services or lose control over their accounts due to device loss or theft.
Innovation Solution
A method to identify inactive user subscriptions by classifying them into groups based on criteria such as duration of inactivity and context factors, matching them with unavailable user data to determine the likelihood of inactivity, and taking appropriate actions such as restricting or archiving accounts to prevent unauthorized access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If zombie accounts and devices are left active, then system infrastructure resources continue to be used, but this creates privacy risks and security vulnerabilities
Solution Approach 1:
The system performs preliminary detection and classification of zombie subscriptions before they can cause security breaches or resource waste. By proactively identifying inactive accounts, apps, and devices through multiple criteria (inactivity duration, context factors, unavailable user data), the system takes preventive action to secure these subscriptions before they pose threats, rather than reacting after incidents occur
Solution Approach 2:
The patent segments zombie subscriptions into three distinct groups based on inactivity likelihood (low, moderate, high), allowing differentiated handling strategies. This segmentation enables the system to apply appropriate security measures and resource management policies to each group, optimizing both security effectiveness and resource efficiency without treating all inactive subscriptions uniformly
2Measurement precision
If comprehensive criteria are used to detect inactive subscriptions, then detection accuracy improves, but system complexity increases
Solution Approach 1:
The detection system is segmented into multiple independent criteria modules, each evaluating a specific aspect (inactivity duration, context factors, unavailable user data). This modular segmentation allows the system to achieve comprehensive detection accuracy while maintaining manageable complexity through organized, independent evaluation components that can be processed systematically
Solution Approach 2:
Different criteria are applied with varying weights and thresholds based on their specific contribution to detecting different types of zombie subscriptions. The system applies local quality assessment by tailoring the detection approach to specific contexts (e.g., different inactivity thresholds for different app types), improving overall detection precision without requiring a uniformly complex system across all scenarios
3Object-affected harmful factors
If zombie subscriptions are identified and classified, then privacy risks are reduced, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary classification of subscriptions into risk groups before implementing security measures or detailed investigations. This preliminary action reduces processing time by quickly categorizing subscriptions based on readily available criteria, allowing high-risk subscriptions to be prioritized for immediate security actions while lower-risk ones receive deferred or reduced attention
Solution Approach 2:
The system applies partial action by focusing comprehensive detection and classification efforts on subscriptions that meet specific risk thresholds, rather than uniformly processing all subscriptions with equal depth. This allows the system to reduce privacy risks effectively while minimizing processing time by applying intensive analysis only where necessary
4Adaptability or versatility
If multiple classification groups are created for inactive subscriptions, then targeted security actions can be applied, but system complexity increases
Solution Approach 1:
The classification system is segmented into three distinct groups (low, moderate, high inactivity likelihood) with clear defining criteria for each. This segmentation provides adaptability by enabling differentiated security responses while maintaining simplicity through well-defined boundaries and systematic classification rules that prevent arbitrary complexity
Data Source
AI summary
A computer-implemented method, including identifying user subscriptions that meet at least one criterion of a plurality of criteria for detecting subscriptions that are no longer active, classifying the user subscriptions into a first group and a second group, where user subscriptions in the first group have at least one context factor associated with the at least one criterion, matching the user subscriptions of the second group with data of unavailable users to produce a third group of user subscriptions, where the user subscriptions of the third group are successfully matched with some of the data of unavailable users.