User Quality Estimation via Terminal Log Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods fail to accurately estimate user-experienced quality of applications, particularly in dynamic content scenarios, due to inability to consider terminal processing time and detect user operations within browser or client software that do not accompany communication with the server, leading to inaccurate quality estimation and difficulty in identifying communication-related quality issues.
Innovation Solution
A user-experienced quality estimation apparatus that utilizes log data for time-series analysis of network, scripting, and rendering processes to estimate waiting time, including terminal processing time, and allows for external monitoring of user-experienced quality by abstracting process types, enabling accurate estimation and bottleneck determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional response time measurement methods are used (HTTP message exchange time), then network communication quality can be monitored, but terminal processing time is not captured leading to inaccurate user-experienced quality estimation
Solution Approach 1:
The patent segments the user-experienced waiting time into distinct components: terminal processing time (script execution, rendering) and network communication time. By separating these measurements and collecting them independently through log data, the system captures complete quality information without losing terminal processing details.
Solution Approach 2:
The patent introduces log data as an intermediary mechanism that bridges the terminal and external monitoring systems. The log file records terminal processing events with timestamps, serving as a mediator that enables external systems to accurately measure terminal processing time without interfering with actual user operations.
2Reliability
If external monitoring of user operations is implemented, then communication-related quality issues can be identified, but user operations within browser or client software that do not accompany communication cannot be detected
Solution Approach 1:
The terminal system performs self-monitoring by automatically generating log entries for its own processing events. The browser or client software itself records script execution start/end times and rendering completion times, enabling comprehensive detection of all user operations including those that occur locally without network communication.
3Measurement precision
If detailed log data collection is implemented for accurate waiting time estimation, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent extracts only the essential timing information needed for quality estimation into a simplified log format. Instead of collecting comprehensive system state data, it selectively records script execution start/end timestamps and rendering completion timestamps, reducing data complexity while maintaining measurement precision.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A user experienced quality estimation apparatus for estimating a user experienced waiting time of an application in an user terminal, includes: a data reception unit configured to obtain, as estimation target logs, a duration time of each process of data obtaining, script execution and screen rendering performed in an estimation target period in the user terminal, and to store the estimation target logs in a received log storage unit; a log exclusion unit configured to read out the estimation target logs from the received log storage unit, and output logs from which logs having data duration time shorter than a predetermined short time process threshold or longer than a predetermined long time process threshold are excluded; a quantization unit configured to calculate, as multiplicity, the number of processes in a timeslot of a constant time length for the logs output from the log exclusion unit; and a continuous region extraction unit configured to extract a continuous region from data quantized by the quantization unit.