Predictive Analysis for Communication Network Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing predictive analytics methods for communication networks are inadequate for information-poor nodes, as they rely on detailed technical attributes that are often missing, leading to poor prediction modeling capabilities, especially for prepaid subscribers where only communication event information is available.
Innovation Solution
A computer-implemented method that enriches information sets for information-poor nodes by assigning them to groups based on communication event information from information-rich nodes, using membership weights as additional features for prediction modeling, thereby optimizing predictive tasks without assuming the reconstruction of missing information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional predictive analytics methods are used that rely on detailed technical attributes, then prediction accuracy is improved for information-rich nodes, but prediction capability deteriorates for information-poor nodes where such attributes are missing
Solution Approach 1:
The patent segments nodes into two categories: information-rich nodes (first set) with detailed technical attributes and information-poor nodes (second set) with limited attributes. This segmentation allows different processing approaches for each group, enabling accurate predictions for both types by treating them differently rather than applying a uniform method that would fail for information-poor nodes
Solution Approach 2:
The patent introduces communication event information as an intermediary element that bridges the gap between information-rich and information-poor nodes. By using this intermediate data type that is available for all nodes regardless of their information richness, the system enables prediction modeling for information-poor nodes without requiring detailed technical attributes that would otherwise be necessary for accurate predictions
2Loss of information
If detailed technical attributes are collected for all nodes, then comprehensive analytical insight is improved, but data collection complexity and resource requirements increase
Solution Approach 1:
The patent applies local quality by collecting detailed technical attributes only for information-rich nodes where such data is available and meaningful, while using a different approach (communication event information) for information-poor nodes. This localized approach to data collection maintains comprehensive analytical insight for the entire network while avoiding the complexity and resource requirements of uniformly collecting detailed attributes for all nodes
3Measurement precision
If separate prediction models are created for information-rich and information-poor nodes, then prediction accuracy for each group is improved, but system complexity increases
Solution Approach 1:
The patent segments the prediction process into two separate models: one for information-rich nodes using detailed technical attributes and another for information-poor nodes using communication event information and group membership features. This segmentation improves prediction accuracy for each node type by using appropriate features for each group, while the modular structure keeps system complexity manageable through clear separation of concerns
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method for carrying out predictive analysis relating to nodes of a communication network. The method comprises the steps of providing communication event information for a first set of nodes and a second set of nodes of the communication network, providing a set of attributes for the nodes of the first set,using said attributes and said communication event information for determining a set of groups among the first set of nodes,assigning each node of the second set to at least one group of the set of groups based at least on the communication event information available for the second group, the assigning resulting in membership information of the nodes of the second set as well as deriving or applying a prediction model for the second set of nodes based on the communication event information for the second set and the membership information.