Hybrid Clustered Prediction Models for Node Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional software solutions for predicting future attributes of nodes in computer modeling are inefficient and inaccurate due to their holistic approach, requiring heavy computing power and causing service interruptions during reconfiguration, which is costly and time-consuming.
Innovation Solution
A hybrid method that generates and executes specific prediction models for node clusters, using clustering algorithms to segment nodes and generate corrective models that improve accuracy without reconfiguring existing software, minimizing service interruptions and computing power usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional software solutions generate a time-series prediction model for all nodes using a holistic approach, then the model can predict an overall trend of all nodes, but the results are not accurate for all node segmentations because they do not capture peculiar attributes of each node or segment
Solution Approach 1:
The patent divides the set of nodes into multiple clusters based on similar attributes. Instead of treating all nodes uniformly, the system segments them into homogeneous groups where each cluster shares common characteristics. This segmentation allows the prediction system to capture peculiar attributes of each node segment while maintaining computational efficiency through clustered modeling.
2Measurement precision
If conventional software solutions generate a time-series model for each node to achieve more accurate results, then the prediction accuracy improves, but heavy computing power is required because a central server must generate and execute a multitude of computer models
Solution Approach 1:
The patent merges nodes with similar attributes into clusters and generates a single prediction model for each cluster rather than individual models for each node. This combining approach maintains prediction accuracy by capturing segment-specific attributes while dramatically reducing computing power requirements by executing fewer clustered models instead of numerous individual node models.
3Measurement precision
If existing prediction models are reconfigured to improve accuracy, then the prediction results can be improved, but the reconfiguration is timely, costly, and creates service interruptions
Solution Approach 1:
The patent performs clustering and model generation in advance before predictions are needed. By pre-segmenting nodes into clusters and generating prediction models for each cluster beforehand, the system avoids time-consuming reconfiguration during operation. This preliminary action ensures prediction accuracy is improved while minimizing service interruptions, as the clustered models are ready for immediate execution.
Data Source
AI summary
Disclosed herein are systems and methods to efficiently execute predictions models to identify future values associated with various nodes. A server retrieves a set of nodes and generates a primary prediction model using data aggregated based on all nodes. The server then executes various clustering algorithms in order to segment the nodes into different clusters. The server then generates a secondary (corrective) prediction model to calculate a correction needed to improve the results achieved by executing the primary prediction model for each cluster. When a node with unknown/limited data and attributes is identified, the server identifies a cluster most similar the new node and further identifies a corresponding secondary prediction model. The server then executes the primary prediction model in conjunction with the identified secondary prediction model to populate a graphical user interface with an accurate predicted future attribute for the new node.