AI Root Cause Change Detection Through Event Record Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The vast amount of data generated in computing environments makes it difficult to efficiently detect root causes of incidents, as manual and existing automated methods struggle to sift through and correlate relevant information effectively.
Innovation Solution
A system utilizing artificial intelligence (AI) processes event and alert records, extracts relevant data values, correlates them to generate incident records, and determines root cause changes, initiating mitigation actions based on AI-driven vector analysis and large language models (LLMs) for root cause analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual methods are used to sift through data, then measurement precision can be maintained, but productivity decreases significantly
Solution Approach 1:
The patent replaces manual data analysis with an AI-based system that uses machine learning models to automatically process event records, detect patterns, and identify root causes. The system substitutes human cognitive processing with automated algorithms that can handle large volumes of data at high speed while maintaining detection accuracy through trained models that recognize causal relationships.
Solution Approach 2:
The patent introduces an AI intermediary layer between raw data and human analysts. This intermediary processes and pre-processes data by filtering, correlating, and prioritizing events based on learned patterns, thereby reducing the information overload for human operators while maintaining the precision of root cause identification through the intermediary's intelligent processing.
2Productivity
If automated methods are used to process data, then productivity increases, but measurement precision decreases
Solution Approach 1:
The patent implements feedback mechanisms where the AI system continuously learns from actual incident resolutions and adjusts its detection models. The system receives feedback when root causes are confirmed or corrected, allowing it to refine its patterns and improve precision over time while maintaining high processing speeds through automated iterative learning and adaptation.
Solution Approach 2:
The patent dynamically adjusts processing parameters based on data characteristics and incident context. The system changes detection sensitivity, correlation thresholds, and analysis depth based on the specific event patterns being analyzed, allowing it to optimize both speed and precision by adapting parameters to the particular incident scenario rather than using fixed parameters.
3Measurement precision
If more data is collected for analysis, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent extracts and focuses analysis on the most critical data elements and patterns relevant to root cause detection. Rather than processing all available data uniformly, the system identifies and extracts only the pertinent features, events, and metadata that are most likely to reveal causal relationships, thereby reducing the effective data volume for analysis while maintaining high precision through targeted extraction of meaningful information.
Solution Approach 2:
The patent segments the complex data analysis task into distinct processing stages: data collection, pattern recognition, correlation analysis, causal inference, and root cause identification. Each stage handles a specific aspect of data processing with dedicated algorithms and models, breaking down the overall complexity into manageable segments that can be processed independently and efficiently.
4Measurement precision
If comprehensive data processing is performed, then root cause analysis accuracy improves, but loss of time increases
Solution Approach 1:
The patent performs preliminary actions by continuously collecting, indexing, and pre-processing event records before incidents occur. The system maintains pre-processed data structures, correlation models, and pattern templates that are ready for immediate use during incident response, eliminating the need for time-consuming data preparation during the actual incident and enabling rapid root cause identification.
Solution Approach 2:
The patent skips unnecessary data processing steps by using AI models to directly identify causal relationships without exhaustive analysis of all possible data combinations. The system rushes through the analysis process by leveraging pre-learned patterns and correlations that point directly to root causes, eliminating tedious manual investigation and intermediate analysis steps that would consume time.
Data Source
AI summary
A system and method for root cause analysis in incident processing, including root cause change detection, is presented. The method includes processing a plurality of event records, including a plurality of alert records and a plurality of change records, each event record generated based on an event in a computing environment; parsing each alert record based on a predetermined data field; extracting from each predetermined data field a data value; correlating a group of alert records of the plurality of alert records based on at least an extracted data value; generating an incident data record based on the extracted data values of the correlated group of alert records; detecting a change record of the plurality of change records related to the incident data record; determining that the change record is a root cause change of the incident data record; and initiating a mitigation action based on the root cause change.