Machine Learning Pipeline for Cloud Incident Entity Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing cloud incidents is challenging due to the large size of unstructured information related to these events, which complicates automated triaging and incident diagnosis.
Innovation Solution
A method and system utilizing a machine learning pipeline to convert cloud incident information into formatted data, recognizing entity names and values associated with potential service impacts, employing unsupervised knowledge extraction and multi-task deep learning models to improve precision and efficiency in incident management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional manual methods are used to process cloud incident information, then information accuracy can be maintained through human review, but processing time and operational complexity increase significantly
Solution Approach 1:
The patent replaces manual mechanical processing of incident information with an automated machine learning system. The ML pipeline automatically extracts entities, classifies incidents, and generates summaries without human intervention, thereby reducing processing time while maintaining accuracy through algorithmic consistency.
Solution Approach 2:
The system enables self-service processing where the incident management system automatically processes and analyzes incident information without requiring manual human review. The automated entity recognition and classification systems perform the work that would traditionally require human analysts.
2Measurement precision
If manual processing methods are used for cloud incidents, then complex unstructured information can be reviewed for accuracy, but operational complexity and resource requirements increase
Solution Approach 1:
The patent replaces complex manual operational processes with automated machine learning systems. The ML pipeline handles entity extraction, classification, and summarization automatically, reducing operational complexity while maintaining information accuracy through consistent algorithmic processing.
Solution Approach 2:
The machine learning system acts as an intermediary between raw incident data and human operators. It pre-processes and structures unstructured incident information into standardized formats with extracted entities and classifications, making subsequent human review more efficient and less complex.
3Productivity
If automated processing systems are implemented, then processing speed and efficiency improve, but accuracy may decrease due to handling of unstructured information
Solution Approach 1:
The system performs preliminary actions by pre-processing incident data through entity extraction and classification before final analysis. The ML pipeline prepares structured representations of unstructured incident information in advance, improving both processing efficiency and accuracy by organizing data beforehand.
Solution Approach 2:
The system implements feedback mechanisms where the ML model continuously learns from processed incidents and improves its accuracy over time. The automated system refines its entity recognition and classification based on feedback from incident outcomes, maintaining high accuracy while preserving processing efficiency.
Data Source
AI summary
Systems and methods for automatic recognition of entities related to cloud incidents are described. A method, implemented by at least one processor, for processing cloud incidents related information, including entity names and entity values associated with incidents having a potential to adversely impact products or services offered by a cloud service provider is provided. The method may include using at least one processor, processing the cloud incidents related information to convert at least words and symbols corresponding to a cloud incident into machine learning formatted data. The method may further include using a machine learning pipeline, processing at least a subset of the machine learning formatted data to recognize entity names and entity values associated with the cloud incident.