Wireless Analytics Data Source Rating for Prediction Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing wireless communications networks face challenges in ensuring the accuracy of machine learning model predictions due to mismatches between training and inference data, particularly when using untrusted or unreliable data sources, which current solutions fail to address effectively.

Innovation Solution

A data analytics function that determines a rating of data sources based on supplementary data to detect and correct analytics data, using local estimation, consumer feedback, or AF-provided weights to improve the selection of data sources and enhance prediction accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If machine learning models use data from multiple data sources for analytics, then the coverage and applicability of analytics services are improved, but the accuracy of predictions deteriorates due to data quality variations and distribution mismatches

Engineering Contradiction:

Improveanalytics service coverageVSAvoidprediction accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent implements data source-specific quality ratings that allow each data source to be evaluated and weighted individually based on its own characteristics, reliability, and data quality. This enables the system to maintain high prediction accuracy by selectively using high-quality data sources while still providing comprehensive analytics coverage through multiple sources.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system dynamically adjusts the rating parameters of data sources based on observed data quality metrics, distribution characteristics, and performance feedback. By changing the weighting parameters adaptively, the system optimizes prediction accuracy while maintaining versatility in data source utilization.

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If data sources are selected without quality assessment, then the system complexity is reduced and ease of operation is improved, but the reliability of analytics data deteriorates due to inclusion of unreliable data sources

Engineering Contradiction:

Improvedata source selection simplicityVSAvoidanalytics data quality

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system performs preliminary quality assessment and rating of data sources before they are used for analytics. By pre-evaluating data source reliability, accuracy, and suitability, the system ensures high data quality without requiring complex real-time selection decisions, thus maintaining ease of operation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms that continuously monitor analytics performance and data quality metrics, using this information to update data source ratings. This automated feedback loop maintains high reliability without requiring manual intervention, preserving operational simplicity.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If data source ratings are determined using supplementary data and feedback mechanisms, then the accuracy of analytics is improved, but the device complexity and processing requirements increase

Engineering Contradiction:

Improveanalytics accuracyVSAvoidrating determination system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system implements a tiered rating approach where not all data sources require full supplementary data analysis. Instead, the system applies appropriate levels of assessment based on data source criticality and available information, achieving high accuracy without the complexity of exhaustive evaluation for every source.

Inventive Principle:

Principle #16Partial or excessive action

4Reliability

If the system continuously monitors and updates data source ratings, then the reliability of analytics data is improved, but the loss of time and processing overhead increase

Engineering Contradiction:

Improvedata quality consistencyVSAvoidrating update time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system implements periodic rating updates triggered by specific events such as data quality threshold violations, performance degradation detection, or scheduled intervals. This event-driven approach maintains data quality reliability while minimizing unnecessary processing and time loss compared to continuous monitoring.

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS20260039560A1Improved accuracy of analytics in a wireless communications network

Publication Date: 2026.02.05 LENOVO (SINGAPORE) PTE LTD

AI summary

There is provided a data analytics function comprising a processor and a receiver. The processor is arranged to generate analytics data for an analytics service using at least one data source. The receiver is arranged to receive an event related to the analytics service. The processor is further arranged to determine a rating of the at least one data source, the rating based on supplementary data.