User Identification via Fingerprint Vector Reliability Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online advertising technologies face challenges in accurately identifying users and associating them with their historical data, especially when cookies expire or are deleted, leading to disassociation of user event and profile data.
Innovation Solution
A system utilizing distributed processors to manage user data, where historical user data is used to determine the most probable user identifier from multiple associated identifiers based on event frequency and recency, with optional parametric modeling to estimate joint probability of frequency and recency values for accurate user detection and targeted advertising.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cookie-based user identification is used, then user identification accuracy is improved, but user identification reliability deteriorates when cookies expire or are deleted
Solution Approach 1:
The system performs preliminary actions by collecting and storing multiple potential user identifiers and their associated historical data before cookie expiration occurs. This includes gathering identifiers from various sources (cookies, device fingerprints, login information) and pre-calculating their reliability scores based on historical accuracy, so that when cookie-based identification fails, the system already has alternative identifiers ready to use immediately without interruption to user tracking or advertising delivery.
2Adaptability or versatility
If multiple user identifiers are associated with a fingerprint vector, then user identification flexibility is improved, but user identification precision deteriorates
Solution Approach 1:
The system changes the parameter of identifier selection by dynamically calculating and comparing reliability scores for each associated user identifier. Instead of treating all identifiers equally, the system evaluates historical data including past identification accuracy, recency of activity, and consistency with current session behavior. This parameter change transforms the selection process from a simple multi-choice scenario into a scored ranking system, where the identifier with the highest reliability score is selected, thereby maintaining precision even with multiple associated identifiers.
Data Source
AI summary
Methods and apparatus for identifying on-line users for advertisement or content targeting are disclosed. Historical user data is obtained in association with user identifiers, which have been unambiguously determined. The historical user data includes event data for one or more on-line user events that have occurred for each user identifier. The historical user data also specify fingerprint vectors of characteristic values that are each associated with specific ones of the user identifiers. A current one of the fingerprint vectors that is ambiguously associated with two or more user identifiers is received. A first user identifier is selected from the associated two or more user identifiers of the current fingerprint vector based on the event data of the historical user data. The selected first user identifier is provided to a server configured to provide advertisement or content based on user profile data that is obtainable for such selected first user identifier.