Network Event Vector Integrity Scoring for KPI Confidence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network and customer experience monitoring systems lack automated processes for providing data integrity measures directly with Key Performance Indicators (KPIs), requiring manual correlation across multiple systems and lacking confidence intervals for users to assess metric significance before taking network or business actions.
Innovation Solution
A method that calculates a presence score and accuracy score for network events by comparing vectors from a telecommunication network testing system with those observed by a monitoring system, using dimensions like IMEI, IMSI, and values such as session length and latency, to display scores visually or textually, enabling users to understand KPI integrity and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual correlation methods are used to assess data integrity between monitoring systems, then users can obtain some confidence in KPI accuracy, but the process requires significant manual effort and time
Solution Approach 1:
The system automatically performs data integrity assessment by having the monitoring system self-correlate its observed network events with expected event characteristics. The integrity score is calculated autonomously without requiring manual intervention, yet still provides comprehensive confidence metrics for KPI accuracy.
Solution Approach 2:
An intermediary correlation engine is introduced that automatically matches and compares network events between different monitoring systems or between observed and expected events. This intermediary component handles the complex correlation logic, freeing users from manual assessment while providing detailed integrity scoring.
2Reliability
If comprehensive data integrity scoring is implemented for all KPIs, then users gain confidence in metric significance, but system complexity increases
Solution Approach 1:
Data integrity scoring is applied locally to individual KPIs and event types rather than uniformly across the entire system. Each KPI receives an integrity score based on its specific correlation results, allowing targeted reliability assessment without requiring complex system-wide changes. The scoring can be applied selectively based on user needs.
Solution Approach 2:
The system changes the parameter of integrity assessment from a complex multi-system correlation process to a simplified scoring metric based on event presence and correlation results. By transforming the complexity into a manageable score parameter, the system provides reliable KPI confidence without requiring complex infrastructure.
3Productivity
If automated integrity scoring is implemented, then operational efficiency improves, but the requirement to correlate multiple data sources increases processing load
Solution Approach 1:
The system extracts only the essential correlation information needed for integrity scoring - specifically whether expected events are present in observed data and basic matching attributes. By taking out only the necessary correlation elements rather than processing all possible data comparisons, the system achieves automated efficiency without excessive processing load.
Solution Approach 2:
The system performs partial correlation - focusing on the most critical event attributes and correlation dimensions needed for meaningful integrity assessment. Rather than exhaustively comparing all possible event parameters, it performs sufficient correlation to generate reliable integrity scores, balancing automation benefits with processing requirements.
Data Source
AI summary
Systems and methods for data integrity scoring and visualization for network and customer experience monitoring are described. In some embodiments, a method may include receiving a first set of vectors, each vector representing a network event generated by a network testing system, each vector including a plurality of dimensions and a first plurality of values, each value associated with a corresponding one of the dimensions. The method may also include identifying a second set of vectors representing at least a portion of the network events as observed by a network monitoring system, each vector in the second set of vectors including the plurality of dimensions and a second plurality of values. The method may further include calculating a presence score as a ratio between a number of vectors in the second and first sets of vectors, and/or an accuracy score as a measure of a discrepancy between corresponding values.