Network Root Cause Analysis via Rules Mining Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying the root cause of operational issues in network devices is challenging due to the large number of features involved, making it difficult to resolve network failures effectively.
Innovation Solution
A method involving a two-phase process: encoding phase to standardize network device features and a causal feature identification phase using association rule mining algorithms like apriori, to generate a standardized database that identifies subsets of features likely causing operational issues, facilitating self-healing mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional manual analysis methods are used to identify root causes, then analysis precision may be maintained through expert judgment, but productivity is severely reduced due to the time-consuming nature of manually examining large numbers of features
Solution Approach 1:
The patent replaces manual expert analysis (mechanical human cognition) with automated machine learning algorithms. The system uses trained models to automatically process and analyze network device features, substituting human experts with computational systems that can handle large volumes of data without manual intervention, thereby dramatically improving productivity while managing complexity through algorithmic processing
Solution Approach 2:
The patent introduces an intermediary layer of feature selection and preprocessing mechanisms between the raw network device data and the root cause analysis. This intermediary layer includes feature importance calculation, dimensionality reduction, and data normalization steps that simplify the complex feature set before feeding it to the analysis model, making the overall system more manageable and productive
2Measurement precision
If all network device features are analyzed in detail, then measurement precision of root cause identification is improved, but loss of time increases due to the extensive processing required
Solution Approach 1:
The patent extracts and selects only the most relevant features from the complete set of network device features using feature importance metrics and selection algorithms. By taking out only the critical features that contribute most to root cause identification, the system maintains high measurement precision while significantly reducing the time required for analysis, as fewer features need to be processed
Solution Approach 2:
The patent applies partial action by analyzing a subset of the most important features rather than all features. The system calculates feature importance scores and focuses computational resources on the top-ranked features, achieving sufficient precision for root cause identification without the excessive time cost of analyzing every single feature in detail
3Productivity
If the number of features examined is reduced, then productivity is improved through faster processing, but measurement precision deteriorates due to potential omission of critical diagnostic information
Solution Approach 1:
The patent performs preliminary action by pre-calculating feature importance metrics, pre-processing data, and pre-selecting relevant features before the actual root cause analysis. This preliminary preparation ensures that when the analysis is executed, only the most critical features are considered, maintaining measurement precision while enabling fast processing during the actual diagnostic operation
Solution Approach 2:
The patent changes parameters by transforming raw network device features into standardized, normalized forms with assigned importance weights. This parameter transformation allows the system to process fewer features with higher confidence, as each feature is optimized and weighted according to its diagnostic value, thereby maintaining precision while improving productivity through efficient parameter management
Data Source
AI summary
In general, embodiments relate to a method, for managing a network, that includes determining an occurrence of an operational issue on a network device of the network, based on the determining, executing an encoding phase and a causal feature identification phase on a feature database, wherein the feature database is associated with the operational issue, identifying a plurality of potential root causes using the encoding phase and the causal feature identification phase, and performing an action based on the plurality of potential root causes.