AI Performance Prediction via Dynamic Model Rebuilding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current monitoring systems for computing and communication infrastructure are static and fail to adapt to real-time changes, leading to inaccurate predictions and inability to handle new hardware, software, or network enhancements, resulting in inefficient system maintenance and performance optimization.
Innovation Solution
An AI-based performance prediction system that uses parallel processing to build and rebuild machine learning models dynamically, selecting the best model for current data and continuously improving with feedback, enabling real-time adaptation and enhanced prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If static monitoring systems are used to track system resources, then the system structure is simple and easy to implement, but the system cannot adapt to real-time changes and new hardware/software enhancements
Solution Approach 1:
The patent implements dynamic model rebuilding where machine learning models are continuously updated in real-time based on incoming data streams. The system transitions from static monitoring to dynamic adaptation by automatically rebuilding models when new patterns are detected, allowing the system to adapt to changing conditions without manual intervention.
Solution Approach 2:
The system performs self-updating through automatic model rebuilding mechanisms. When the system detects new hardware, software, or network configurations, it autonomously rebuilds its predictive models without requiring external reconfiguration, enabling self-adaptation to evolving infrastructure.
2Measurement precision
If traditional monitoring systems are used, then the implementation is straightforward, but prediction accuracy deteriorates over time as system changes occur
Solution Approach 1:
The system performs preliminary model rebuilding in the background before predictions are needed. By continuously training models on incoming data streams and maintaining multiple model versions, the system ensures accurate predictions are ready when needed without causing time delays during actual prediction operations.
Solution Approach 2:
The patent implements continuous model training and rebuilding processes that operate in parallel with normal monitoring operations. This continuous action ensures prediction accuracy is maintained over time without interrupting the monitoring function, as models are constantly refined based on new data.
3Reliability
If machine learning models are rebuilt frequently to adapt to changes, then prediction accuracy improves, but computational resources and processing time increase
Solution Approach 1:
The system applies partial model rebuilding by selectively updating only the portions of models that need adjustment based on detected changes. Rather than completely rebuilding all models frequently, the system performs targeted updates on specific model components, reducing computational overhead while maintaining prediction reliability.
Solution Approach 2:
The patent implements periodic model rebuilding with variable intervals based on detected system stability. When the system operates normally, model rebuilding occurs at extended intervals to conserve resources. When changes are detected, the rebuilding frequency increases automatically, creating an adaptive periodic action pattern.
Data Source
AI summary
An Artificial Intelligence (AI) based performance prediction system predicts the performance and behavior of an entity via a complex structure made of iterative and parallel machine learning (ML) model rebuilds with real time data collection. The engine selects a best model at every level and scores the entity to help in predicting the behavior of the entity. Model selection is based on various model selection criteria. The selected model determines a propensity score that indicates a likelihood of the entity migrating from a currently categorized segment to another segment of higher or lower value. Accordingly, messages or alerts with one or more of corrective actions or system enhancements can be transmitted based on the status of the entity via various targeting channels and a post treatment analysis is carried out to find the effect of the corrective actions on the entity. The feedback from the entity in response to the implemented corrective actions or system enhancements is collected for further training the AI based model.