Machine Learning Model Selection for Proactive Telecommunications Network Care
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual network monitoring and management become inefficient as network activity increases, leading to delayed processing of trouble tickets and alarms, resulting in service outages and unsatisfactory customer experiences due to the inability to identify issues before customer notifications are generated.
Innovation Solution
A customer care system that trains machine learning models with anonymized historical telecommunications data to select an optimum model for predicting customer care actions, which are then applied proactively to the network, decoupling training from prediction to enhance data security by maintaining customer data on-site.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual network monitoring and management are used, then network activity can be monitored, but processing efficiency decreases and response time increases as network activity increases
Solution Approach 1:
The patent replaces manual network monitoring and management with an automated machine learning-based system. The ML model automatically analyzes network data, identifies issues, and determines customer care actions, substituting human operators with an intelligent automated system that processes data faster and more efficiently.
Solution Approach 2:
The system enables self-service by allowing the network monitoring system to automatically identify issues and determine appropriate customer care actions without requiring constant human intervention. The ML model autonomously processes network data and generates actionable insights, reducing dependency on manual processing.
2Measurement precision
If machine learning models are trained with customer data at remote servers, then model accuracy can be improved, but data security and privacy are compromised
Solution Approach 1:
The patent extracts the model training process from the remote server environment and relocates it to the local network edge. By training models locally using anonymized historical data, the system maintains model accuracy while eliminating the security risks associated with transmitting and storing sensitive customer data on remote servers.
Solution Approach 2:
The system implements local quality by processing and training machine learning models at the local network edge rather than centralizing data processing at remote servers. This distributed approach allows each location to maintain its own data security while still benefiting from advanced analytics capabilities.
Data Source
AI summary
In some implementations, a device may receive, from a monitoring device, telecommunications data associated with a telecommunications network. The device may train a plurality of machine learning models with the telecommunications data to generate a plurality of trained machine learning models. The device may generate accuracy scores for the plurality of trained machine learning models based on training the plurality of machine learning models. The device may select an optimum machine learning model based on several indicators, such as the accuracy, precision, and/or the like. The device may provide the optimum machine learning model to the monitoring device associated with the telecommunications network. The optimum machine learning model may cause the monitoring device to process real time telecommunications data of the telecommunications network, with the optimum machine learning model, to determine a customer care action, and may cause the customer care action to be implemented in the telecommunications network.