Automated Identity Detection via User Judgment Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social networks face challenges in detecting automated identities, as existing methods like asking users to enter noisy or distorted text have become less effective due to improved OCR techniques, allowing fake identities to gather data for nefarious purposes.
Innovation Solution
A computer-implemented method that displays representations of two identities to users, along with a choice indication mechanism, allowing users to judge which identity appears more or less automated, and determines the likelihood of automation based on user judgments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If OCR-based text verification is used to detect automated identities, then the ability to distinguish human users from bots is improved, but the effectiveness deteriorates as OCR techniques improve
Solution Approach 1:
Instead of asking users to prove they are human by performing tasks that bots struggle with (OCR), the system inverts the approach by having users perform tasks that bots excel at (comparing visual profiles) while bots struggle to distinguish real from fake profiles. This reversal exploits the complementary strengths and weaknesses of human and automated systems.
Solution Approach 2:
The system introduces an intermediary verification mechanism where users act as mediators to verify other users' identities through profile comparison. This intermediary human judgment layer adds a verification step that is difficult for automated systems to bypass, as it requires subjective human assessment of profile authenticity.
2Measurement precision
If human judgment is used to identify automated identities, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The system implements self-service verification where users voluntarily participate in verifying other profiles by comparing them. This user-driven approach eliminates the need for complex automated verification systems, as the verification capability is provided by the users themselves through their natural ability to assess profile authenticity.
Solution Approach 2:
The system merges the identity verification function with the existing social networking profile viewing experience. By combining profile comparison with the natural user behavior of viewing and evaluating profiles, the system integrates verification into the existing workflow without adding separate complex verification infrastructure.
Data Source
AI summary
A computer-implemented method of detecting automated identities, including. The method includes, using one or more computing devices, providing for display to a user a first representation of a first profile for a first identity and a second representation of a second profile for a second identity, providing for display to the user a choice indication mechanism for indicating a judgment about the first representation of the first profile for the first identity and the second representation of the second profile for the second identity, receiving an indication of the judgment from the user about the first representation of the first profile for the first identity and the second representation of the second profile for the second identity, and determining whether at least one of the first identity, the second identity, or the user is more or less likely to be automated, based at least in part on the received indication.