Service Hotspot Identification Using LLM-Labeled Issue Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
IT teams struggle to identify critical issues among a multitude of incoming requests and incidents in IT service management systems, leading to delays in addressing these issues, which can cause disruptions in business operations and loss of productivity and revenue, due to the lack of a proactive framework for monitoring and analyzing IT infrastructure.
Innovation Solution
A method and system using a causal analysis server that processes raw data through chunking, tokenizing, embedding, clustering, and labeling with a Large Language Model (LLM) to identify domain-specific service management hotspots, enabling proactive identification and mitigation of potential issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If IT teams manually monitor and analyze incoming requests and incidents, then they can identify critical issues, but the sheer volume and diversity of issues overwhelm the teams, leading to delays in addressing critical issues
Solution Approach 1:
The patent segments the overwhelming volume of IT service management data into manageable clusters using unsupervised machine learning algorithms. Similar incidents and requests are grouped together, allowing IT teams to analyze clustered patterns rather than individual items, thereby improving both identification accuracy and response speed.
Solution Approach 2:
The patent introduces an intermediary automated analysis system that acts as a mediator between the raw data and IT teams. This system uses natural language processing and machine learning to pre-analyze, cluster, and prioritize issues before presenting them to human analysts, reducing their cognitive load and enabling faster response to critical issues.
2Reliability
If IT teams adopt a reactive approach to handle issues as they arise, then they can respond to immediate problems, but they miss opportunities to proactively identify and mitigate hotspots, leading to recurrent issues and disruptions
Solution Approach 1:
The patent implements preliminary action by using unsupervised machine learning to continuously analyze and cluster service management data in real-time. This proactive approach identifies emerging patterns and potential hotspots before they manifest as full-blown incidents, allowing IT teams to take preventive measures and maintain service continuity.
Solution Approach 2:
The patent establishes a feedback loop where clustered incident data continuously informs the system about emerging patterns and hotspots. This feedback mechanism enables the system to adapt and identify potential issues before they affect service delivery, transitioning from reactive to proactive issue management.
3Ease of operation
If IT teams lack a structured framework for analyzing IT infrastructure, then they can operate with simple processes, but they struggle to prioritize critical issues and detect emerging problems timely
Solution Approach 1:
The patent transforms unstructured service management data into structured clusters using machine learning parameters and algorithms. By changing the parameters of data organization from simple chronological ordering to similarity-based clustering, the system achieves both operational simplicity and accurate issue prioritization through automated pattern recognition.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
This disclosure relates to method and causal analysis server for identifying domain-specific service management hotspots. The method includes receiving raw data corresponding to plurality of domain-specific issues; generating plurality of chunks from the raw data using an optimal chunking and tokenizing algorithm selected from chunking and tokenizing algorithms to obtain chunked data; creating a plurality of embeddings from the chunked data through an embedding model; clustering the plurality of embeddings into a set of clusters and a set of sub-clusters associated with each of the set of clusters, based on a similarity analysis; determining a set of cluster labels corresponding to the set of clusters and a set of sub-cluster labels corresponding to the set of sub-clusters, via a Large Language Model (LLM); and identifying a plurality of hotspots corresponding to the plurality of domain-specific issues based on the set of cluster labels and the set of sub-cluster labels.