Passive User Assessment in Online Video Advertising
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online video advertisement technologies lack the ability to accurately determine whether video advertisements are displayed to human users or machines, leading to inaccurate billing and potential loss of ad opportunities due to the inability to differentiate between human and automated viewers.
Innovation Solution
A method and apparatus that assess whether a user is a human or machine by analyzing short-term and long-term ad viewing patterns, determining a user's daily schedule, and selectively targeting advertisements based on this information, allowing for passive assessment without requiring user interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video advertisements are delivered to all users without differentiation, then ad coverage is maximized, but billing accuracy deteriorates due to inability to distinguish human users from machines
Solution Approach 1:
The system segments users into different categories (human users vs. machine users) based on their interaction patterns with video advertisements. By analyzing usage patterns and behavior characteristics, the system divides the user base into distinct groups that can be billed differently, thereby improving billing accuracy without requiring complex manual verification processes for each user
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring user interactions with video advertisements and using this information to assess whether users are human or machines. The usage pattern data is fed back into the assessment model to refine user classification over time, enabling accurate billing decisions based on accumulated behavioral evidence
2Loss of energy
If video advertisements are delivered without user assessment, then ad delivery speed is maintained, but resource waste increases due to machine-generated impressions
Solution Approach 1:
The system performs preliminary assessment of users by analyzing their historical usage patterns and interaction behaviors before delivering video advertisements. By pre-evaluating users based on their past behavior with video ads, the system identifies machine users in advance and prevents resource waste by excluding them from future ad deliveries, rather than wasting resources on assessment after each ad delivery
Solution Approach 2:
The system enables users to effectively assess themselves by analyzing their own interaction patterns with video advertisements. The usage data generated by user behavior automatically serves as the assessment input, eliminating the need for external verification processes and reducing both time and resource requirements for user evaluation
3Productivity
If traditional ad targeting methods are used without usage pattern analysis, then campaign setup is simple, but ad effectiveness deteriorates due to inability to target appropriate users
Solution Approach 1:
The system changes the parameters used for ad targeting from basic demographic information to behavioral usage patterns. By analyzing how users interact with video advertisements (viewing duration, engagement level, frequency), the system identifies meaningful patterns that predict ad effectiveness, enabling more productive targeting decisions based on actual user behavior rather than static user profiles
Data Source
AI summary
Whether video advertisements are being delivered to a human user or a machine is determined passively, i.e., without having the user to perform any explicit actions. Based on the time and frequency of ad requests generated from a user device, a daily schedule is estimated for the user. The estimated daily schedule is compared with a daily schedule pattern to determine whether the user is a human user or not.