Dynamic Feature Trigger for Network Data Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network elements face challenges in accurately analyzing data due to changes in the network environment, leading to reduced accuracy in identifying abnormal data patterns.
Innovation Solution
A data processing method that involves receiving a dataset from a network element, extracting data features, and sending trigger information to improve feature type identification, allowing for more accurate analysis of network status despite environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the network element analyzes data based on fixed feature types, then the analysis process is simple, but the accuracy of analyzing data changes when the network environment changes is reduced
Solution Approach 1:
The patent applies dynamics by transitioning from fixed feature types to dynamic feature type determination. The network element receives feature type information from the management device based on actual data characteristics, allowing the analysis system to adapt to changing network environments. This resolves the contradiction by making the feature types flexible rather than static, improving accuracy without requiring overly complex manual configuration.
Solution Approach 2:
The patent changes the parameter of feature types from fixed to variable. The management device determines appropriate feature types based on data characteristics and sends them to the network element. This parameter change allows the system to maintain high accuracy across different network conditions while keeping the analysis process relatively simple through automated parameter adjustment.
2Reliability
If the network element collects and analyzes all data pieces periodically, then comprehensive monitoring is achieved, but the difficulty of analyzing data increases when the network environment changes
Solution Approach 1:
The patent changes the parameter of data analysis from generic to specific by introducing determined feature types. The management device analyzes data characteristics and determines appropriate feature types, which are then used by the network element for focused analysis. This reduces the difficulty of detecting and measuring data changes while maintaining comprehensive monitoring through targeted feature analysis.
Solution Approach 2:
The management device acts as an intermediary between data collection and analysis. It receives data from the network element, determines appropriate feature types based on data characteristics, and sends the determined feature types back to guide the analysis process. This intermediary function reduces the complexity of direct analysis by the network element while maintaining reliable comprehensive monitoring.
3Ease of manufacture
If the network element uses fixed analysis methods, then the implementation is straightforward, but the accuracy of identifying abnormal data patterns decreases when the network environment changes
Solution Approach 1:
The patent implements feedback by having the management device analyze data characteristics, determine appropriate feature types, and send them back to the network element. This feedback loop allows the system to maintain ease of implementation through automated determination while improving the accuracy of identifying abnormal patterns by adapting feature types to actual data characteristics.
Solution Approach 2:
The management device performs preliminary action by determining feature types before the network element conducts detailed analysis. This preliminary determination of appropriate feature types based on data characteristics simplifies the subsequent analysis process while improving accuracy, as the network element only needs to apply the pre-determined feature types rather than complex analysis methods.
Data Source
AI summary
A data processing method includes receiving a first dataset sent by a network element, where the first dataset includes a plurality of pieces of first data obtained by the network element; obtaining, based on the first dataset, at least one data feature corresponding to the plurality of pieces of first data; and sending trigger information to the network element, where the trigger information includes at least one data feature and/or at least one feature type, and at least one feature type is related to at least one data feature.