Virtual Environment User Screening via Fraud Score Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtual environments, it is challenging for advertisers to track viewability of digital advertisements due to dynamic and complex graphics, and there is a need to distinguish between human and non-human users to prevent malicious activities such as bot-generated views.
Innovation Solution
A method for screening users in virtual environments by determining representative values of metrics such as angles between content item orientation vectors and camera view directions, on-screen real estates, view duration, and comparison of view duration with camera control duration, and comparing these values to reference values to determine a fraud score, thereby identifying human users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple metrics are analyzed to determine user authenticity, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system segments the user authentication process into multiple independent metric evaluations: view duration analysis, camera control duration comparison, on-screen real estate measurement, and orientation angle calculation. Each metric is computed separately and then aggregated to form a comprehensive fraud score, allowing precise detection while maintaining modular system architecture that manages complexity.
2Measurement precision
If view duration and camera control duration are compared to detect bots, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system performs preliminary tracking of view duration and camera control duration throughout the user's interaction with the virtual environment. These metrics are continuously recorded and stored before the final authentication decision is made, allowing the comparison to be performed efficiently without adding significant processing time at the decision point.
3Measurement precision
If on-screen real estate and orientation angles are calculated for advertising content, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system introduces intermediary computational elements such as orientation vectors for content items and virtual cameras, and uses these vectors to calculate angles and on-screen real estate. These intermediaries provide a structured mathematical framework that simplifies the calculation process and enables precise measurement of advertising viewability without requiring complex direct geometric computations.
4Reliability
If comprehensive metrics are tracked for each user, then reliability of advertising metrics is improved, but loss of information increases
Solution Approach 1:
The system extracts only the essential and most discriminatory metrics from the vast amount of available user interaction data. Specifically, it focuses on view duration, camera control duration, and key geometric parameters (angles and on-screen real estate) rather than attempting to process all possible interaction data. This selective extraction maintains data integrity for fraud detection while minimizing processing overhead.
Data Source
AI summary
Methods, systems, and media for screening users of a virtual environment are provided. In some embodiments, the method comprises: determining one or more first representative values of one or more first metrics for a first user, wherein the one or more first metrics are associated with one or more content items presented within a virtual environment and wherein the one or more first metrics includes: (i) angles between orientation vectors of one or more content items and view directions of one or more virtual cameras in the virtual environment, wherein the one or more virtual cameras are controlled by the first user; (ii) on-screen real estates of the one or more content items based on distances between the one or more content items and the one or more virtual cameras controlled by the first user; (iii) view duration in which a content item of the one or more content items was at least partially presented for the first user on a rendered screen of the virtual environment during a predetermined amount of time; and (iv) a comparison between the view duration and a duration in which the one or more virtual cameras were controlled by the first user during the predetermined amount of time; in response to determining the one or more first representative values of the one or more first metrics, comparing the one or more first representative values to one or more corresponding reference values, wherein comparing the one or more first representative values to the one or more corresponding reference values comprises determining a fraud score based on the one or more first representative values and the one or more corresponding reference values; in response to determining the fraud score, determining that the fraud score is within a first predetermined range of values; and, in response to determining that the fraud score is within the first predetermined range of values, determining that the first user in the virtual environment is a human user.