Screen Capture for User-Perceived Service Performance Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately measure service performance as perceived by users due to variations in screen loading times and network conditions, and lack of consideration for user environments, making it difficult to compare performance differences between services.
Innovation Solution
A method and system that captures screens based on a time unit until a landing image is loaded, determines loading completion, and indexes performance using loading time and an evaluation reference time to measure user-perceived service experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If internal messages from browser and system are used to measure loading completion time, then measurement can be performed, but the measured time differs greatly from the actual user-perceived completion time
Solution Approach 1:
The patent introduces a screen capture mechanism as an intermediary between the browser's internal loading messages and the actual user perception. By capturing screenshots at regular intervals and comparing them with expected landing images, the system mediates the measurement process to accurately reflect user-perceived loading completion time, resolving the discrepancy between system-reported time and actual user experience.
Solution Approach 2:
The patent replaces the traditional mechanical/browser-based timing mechanism with a visual verification mechanism. Instead of relying on browser internal messages or system timestamps, the system uses screen capture images to verify when the landing image has actually loaded, substituting the abstract timing system with a concrete visual verification system that directly measures user-perceived completion.
2Measurement precision
If screen capture is performed continuously to determine loading completion, then accurate user-perceived performance can be measured, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-defining the expected landing image and setting a time threshold before actual measurement begins. The system captures screens at regular intervals (e.g., every 100ms) but stops when the landing image is detected or the threshold is reached, avoiding continuous capture indefinitely. This preliminary preparation enables accurate measurement while limiting total processing time.
Solution Approach 2:
The patent implements a skipping mechanism where screen captures are performed at optimized intervals rather than continuously. By capturing screens at strategic time points (e.g., every 100ms) and immediately comparing them with the expected landing image, the system rushes through the measurement process efficiently, stopping as soon as the landing image is detected or the time threshold is reached, thus reducing unnecessary processing time.
3Quantity of substance
If performance data is collected without indexing, then raw data is available, but performance differences between services cannot be instantly recognized
Solution Approach 1:
The patent transforms raw performance data into indexed performance metrics by changing the representation parameters. Instead of storing only raw loading times, the system calculates performance indices by dividing the evaluation reference time by the actual loading time, creating a normalized metric that enables instant recognition and comparison of performance differences between services.
Solution Approach 2:
The patent segments the performance evaluation process into distinct components: collecting raw loading time data, calculating performance indices, and comparing services. By segmenting the data processing and creating separate performance indices for different services, the system enables efficient comparison and recognition of performance differences without processing the entire raw data set.
Data Source
AI summary
Methods and systems for evaluating a service experience performance of a user may be provided. A method of evaluating a service experience performance may include capturing a screen based on a time unit until a landing image according to a scenario of a service is loaded, determining that loading of the landing image is completed based on the captured screen, deriving an amount of loading time until the loading of the landing image is completed, and indexing performance for the scenario of the service based on the amount of loading time and an evaluation reference time.