ITSM Incident Clustering for Automated Problem Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current IT service management (ITSM) problem identification from incident reports is a cumbersome manual process that involves manual correlation of multiple tickets, leading to inefficiencies and missed issues due to the reliance on heuristics and rule-based methods, especially when dealing with large volumes of incidents.
Innovation Solution
A computer-implemented method using unsupervised machine learning clustering algorithms to automatically convert textual data from ITSM incident reports into vectors, estimate optimal cluster numbers, and recursively refine clusters based on cluster quality scores, eliminating the need for manual input and ensuring cohesive problem representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual correlation methods are used to identify problems from incident reports, then domain knowledge and heuristics can be applied, but the process becomes time-consuming and many problems go undetected
Solution Approach 1:
The patent replaces manual mechanical analysis with automated machine learning algorithms. The system uses unsupervised learning models to automatically correlate incident reports, eliminating the need for manual spreadsheet analysis while maintaining or improving detection accuracy through consistent application of learned patterns across large datasets.
Solution Approach 2:
The system enables self-service problem identification by automatically processing incident reports without requiring manual intervention. The machine learning models autonomously perform correlation, clustering, and problem identification, allowing the system to serve itself in analyzing large volumes of incident data that would otherwise require human analysts.
2Adaptability or versatility
If manual spreadsheet methods are used for incident analysis, then flexibility in applying heuristics is maintained, but the volume of incidents that can be analyzed is limited
Solution Approach 1:
The patent changes the parameters of analysis by transitioning from manual heuristic evaluation to automated machine learning model parameters. The system adjusts model parameters such as clustering algorithms, similarity thresholds, and feature weights to maintain adaptability while processing vastly larger volumes of incidents than manual methods can handle.
Solution Approach 2:
The machine learning system provides universal problem identification capabilities that work across different incident types and volumes. The same automated pipeline can handle diverse incident reports using learned patterns, making the system versatile and adaptable to various problem domains while maintaining high processing throughput.
3Reliability
If more incidents are analyzed manually, then more problems may be detected, but the time and resources required increase significantly
Solution Approach 1:
The patent replaces complex manual processing resources with automated computational systems. Machine learning models efficiently analyze large incident volumes using algorithmic patterns rather than human cognitive resources, maintaining complete problem detection while reducing the complexity and cost of processing resources required.
Data Source
AI summary
Information technology service management (ITSM) incident reports are converted from textual data to multiple vectors using an encoder and parameters are selected, where the parameters include a base cluster number and a threshold value. A base group of clusters is generated using an unsupervised machine learning clustering algorithm with the vectors and the parameters as input. A cluster quality score is computed for each of the base group of clusters. Each cluster from the base group of clusters with the cluster quality score above the threshold value is recursively split into new clusters until the cluster quality score for each cluster in the new clusters is below the threshold value. A final group of clusters is output, where each cluster from the final group of clusters represents ITSM incident reports related to a same problem.