Machine Learning Engine Segmentation for Mobile Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning in mobile terminal devices faces computational intensity and time-consuming data acquisition and training, with high real-time requirements for prediction results, leading to performance issues and user privacy concerns when models are updated online.
Innovation Solution
A machine learning engine is implemented with a core learning application module and a prediction output service module, where the core learning application module performs machine learning independently and outputs results to the prediction output service module, which buffers and provides the results to other system modules in real-time, allowing for efficient and timely access without direct calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If machine learning is applied to system optimization with high real-time requirements, then prediction speed is improved, but computational complexity and training time increase
Solution Approach 1:
The machine learning system is segmented into two distinct modules: a core learning application module responsible for model training and updates, and a prediction output service module responsible for real-time predictions. This segmentation allows the prediction module to serve requests without undergoing complex training operations, thereby improving prediction speed while maintaining necessary computational complexity in the training module.
2Measurement precision
If data acquisition and training are performed on mobile devices, then model accuracy is improved, but training time and energy consumption increase
Solution Approach 1:
The core learning application module performs model training and updates in advance, preparing optimized models before they are needed for prediction. The trained models are then transferred to the prediction output service module, which can immediately serve real-time prediction requests without performing time-consuming training operations, thus reducing training time while maintaining model accuracy.
3Adaptability or versatility
If models are updated online, then system adaptability is improved, but user privacy security deteriorates
Solution Approach 1:
The invention extracts the model training and update functionality into a separate core learning application module that operates independently from the prediction service. This separation allows the prediction output service module to receive pre-trained models without exposing user data or prediction logic to external networks, thereby maintaining system adaptability through updated models while protecting user privacy security by keeping the prediction process isolated from online threats.
Data Source
AI summary
A method for implementing a machine learning engine and apparatus, a terminal device (50), and a storage medium (60) are disclosed. The apparatus may include: a core learning application module (101) with an independent application process and a prediction output service module (102) located in the system process (100); the core learning application module (101) is configured to output a prediction result generated by machine learning to the prediction output service module (101); and the prediction output service module (102) is configured to buffer the prediction result when receiving the predictions result sent by the core learning application module (101).