Action Sequence Alignment for Crash Cause Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining sequences of actions leading to undesirable events in computer applications, such as crashes, are inefficient and inaccurate, requiring manual analysis of vast amounts of telemetry data.
Innovation Solution
A system and method that aligns and compares action sequences using fitting alignment to identify critical sequences leading to predetermined events by inserting gaps for non-matching data, allowing for rapid identification of common actions contributing to the event.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis of telemetry data is used to determine action sequences leading to crashes, then accuracy in identifying critical sequences is improved, but time consumption and labor requirements increase significantly
Solution Approach 1:
The system performs self-service by automatically analyzing telemetry data and identifying critical action sequences without requiring manual human investigation. The automated algorithm processes vast amounts of data, aligns action sequences, and extracts critical patterns independently, eliminating the need for human analysts to manually review each sequence while maintaining high accuracy in identifying crash-causing actions.
Solution Approach 2:
The patent replaces the mechanical manual analysis process with an automated computational system. The algorithm mechanically processes telemetry data through standardized operations including sequence alignment, gap insertion for non-matching data, and critical sequence extraction, substituting human cognitive processing with automated computational mechanisms that operate continuously without fatigue or subjectivity.
2Measurement precision
If manual review of action sequences is performed, then accuracy in determining crash causes is improved, but productivity and efficiency deteriorate
Solution Approach 1:
The system autonomously performs the entire analysis process from data collection to critical sequence identification without human intervention. The automated algorithm independently reviews action sequences, aligns them using fitting alignment, and extracts critical patterns, thereby maintaining high accuracy while dramatically improving productivity by eliminating manual review requirements.
Solution Approach 2:
The patent changes the operational parameters from manual human analysis to automated computational processing. The system transforms qualitative human review into quantitative algorithmic processing, using parameters such as sequence alignment thresholds, gap insertion rules, and critical sequence extraction criteria to maintain analytical accuracy while achieving high-speed automated processing that significantly boosts productivity.
3Adaptability or versatility
If fitting alignment with gap insertion is applied to action sequences, then ability to handle variable-length sequences is improved, but computational complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the action sequences into discrete, comparable units and processing them through standardized alignment operations. The fitting alignment algorithm segments the sequences into corresponding positions, inserts gaps for non-matching segments, and processes each segment independently, which manages computational complexity while maintaining the ability to handle variable-length sequences flexibly.
Data Source
AI summary
A system and method to collect an actions list of action sequences in an application leading to a predetermined resulting event, create pairs of the action sequences, apply a fitting alignment to the action sequence pairs to create fitted action sequence pairs, wherein non-matching data between fitted action sequences of each pair is replaced with gaps to ensure that the first and second fitted action sequences are of equal length and are aligned with one another with the gaps being located at index positions the fitted action sequences corresponding to index positions of non-matching data, and delete data, for each of the fitted action sequence pairs, corresponding to the gaps to create a critical sequence of actions for each of the fitted action sequence pairs representing, respectively, common actions of the fitted action sequences of each of the fitted action sequence pairs leading to the predetermined resulting event.