Application Link Classification Using Presentation Duration Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately identify and remove broken application links, which can cause application crashes and user frustration, especially when links point to content that is moved, updated, or becomes unavailable, and are not effectively tested by traditional crawlers.
Innovation Solution
A system utilizing machine learning to classify application links as broken or working based on presentation durations, grouping links by prefix patterns, and generating alerts for broken links, thereby reducing network timeouts and improving accuracy in identifying broken links.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional link testing methods are used, then the system is simple to implement, but the accuracy of broken link identification is insufficient
Solution Approach 1:
The patent replaces traditional mechanical link testing methods with a machine learning-based system that uses presentation duration data to classify links as broken or working. The machine learning model processes temporal patterns from user interaction data to automatically identify broken links with higher accuracy than conventional methods.
Solution Approach 2:
The patent introduces presentation duration data as an intermediary feature between user interactions and link status determination. This intermediate measurement serves as the basis for machine learning classification, enabling indirect but accurate identification of broken links through temporal pattern analysis rather than direct link testing.
2Reliability
If broken links are not removed, then the application continues to function, but application crashes occur and user experience deteriorates
Solution Approach 1:
The patent performs preliminary identification and classification of broken links before they cause application crashes. By continuously monitoring presentation durations and training machine learning models to detect broken link patterns in advance, the system can remove or flag problematic links proactively, preventing crashes before they occur.
Solution Approach 2:
The patent implements a feedback mechanism where presentation duration data from user interactions is continuously collected and used to train machine learning models. The model outputs predictions about link status that feed back into the system to automatically remove broken links, creating a closed-loop system that adapts to changing link statuses and prevents crashes.
3Productivity
If manual link testing is performed, then the process is simple, but the scale of link detection is limited
Solution Approach 1:
The patent enables the system to automatically monitor and classify its own links using machine learning models trained on presentation duration data. The system self-services by continuously collecting interaction data, training models, and identifying broken links without requiring manual intervention, thereby scaling to large numbers of links while maintaining operational simplicity through automation.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying broken network connections. In one aspect, a system includes front-end server(s) that receive data specifying, for multiple different user interactions with one or more application links that link to a given application, presentation durations specifying how long application content linked to by the application link was presented following the multiple different user interactions with the application link(s). Back-end server(s) that communicate with the front end server(s) can classify each application link as broken or working based on application of a machine learning model to the presentation durations for the application link. The machine learning model can be generated using labeled training data. The back-end server(s) can generate and output an alert identifying an application link as a broken link based on the application link being classified as broken by the machine learning model.