User Session Partitioning for Active Time Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advertisers face challenges in accurately measuring and comparing user engagement with mobile advertisements, particularly due to varying content lengths and difficulty in determining active time spent, which limits their ability to optimize campaigns and prevent click fraud.
Innovation Solution
A system that divides user sessions into partitions based on predefined schemes, such as event detection or fixed time intervals, to estimate user engagement time, allowing for accurate reporting and optimization of content delivery and fraud prevention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If advertisers use view count metrics to measure advertisement performance, then they can estimate the number of users reached, but they cannot accurately compare different advertisements with varying content lengths or determine actual user engagement time
Solution Approach 1:
The patent segments the continuous user session into discrete time partitions (e.g., 10-second intervals) and assigns engagement states (engaged, disengaged, unknown) to each partition based on detected events. This segmentation allows precise measurement of active time spent by breaking down the overall session into measurable units, directly resolving the inability to accurately measure user engagement with content of varying lengths.
2Reliability
If publishers inflate advertisement view metrics to increase reported performance, then they can appear to reach more users, but they cannot actually deliver genuine user engagement without detection
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring user interactions and device events during advertisement playback, then using this feedback to adjust engagement state assignments for each time partition. This real-time feedback loop detects fraudulent behavior patterns (such as lack of expected user interactions) and adjusts metrics accordingly, making fraud detection possible while maintaining reliable measurement of genuine engagement.
3Productivity
If advertisers want to optimize advertisement campaigns based on user engagement, then they need accurate active time spent data, but current metrics do not provide sufficient information for effective optimization
Solution Approach 1:
The patent performs preliminary actions by pre-defining engagement state thresholds and event detection rules before analyzing user sessions. The system establishes criteria for what constitutes engaged, disengaged, or unknown states in advance, then applies these predetermined rules to automatically classify each time partition. This preliminary setup enables efficient optimization by providing structured, actionable engagement data that advertisers can use to refine their campaigns based on actual user reactions.
Data Source
AI summary
Systems, methods, and computer-readable storage media for optimizing and reporting time spent by a user engaged in a session. The system first obtains data associated with a presentation of an item of content at a mobile device, the presentation being divided into multiple partitions. Based on the data, the system adjusts at least one respective length of time associated with the multiple partitions to yield at least one adjusted length of time, the at least one adjusted length of time reflecting estimated time of user engagement with content associated with at least one of the multiple partitions. The system then determines an amount of time spent by a user at the mobile device engaging in the presentation based on the at least one adjusted length of time and the multiple partitions.