Driver Identification Model Training With Bagging for Better Generalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current driving model identification methods have poor generalization ability and low accuracy due to the use of a single model for identifying whether a mobile phone user is driving, which affects the reliability of the identification results.
Innovation Solution
A method involving the acquisition of training behavior data associated with a user identifier, extraction of training driving data, and the use of a bagging algorithm to train and test a driving model, enhancing the model's generalization and accuracy by distinguishing between driving and non-driving behaviors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single model is used to identify whether a mobile phone user drives, then the identification process is simple, but the generalization ability is poor and identification accuracy is low
Solution Approach 1:
The patent segments the identification process into multiple independent models (first driving model, second driving model, etc.) instead of using a single model. Each model is trained on different datasets and processes specific aspects of driving behavior identification. The final identification result is obtained by combining the outputs of these multiple models, which improves generalization ability and accuracy while maintaining manageable complexity through modular design.
2Device complexity
If a single model is used to identify whether a mobile phone user drives, then the system structure is simple, but the generalization ability is poor
Solution Approach 1:
The patent creates multiple driving models that serve universal functions in the identification system. Each model is designed to handle different aspects of the identification task and can be applied to various user scenarios. The models work together to provide a comprehensive solution that adapts to different driving behaviors and conditions, thereby improving generalization ability without requiring complete system redesign for different cases.
Solution Approach 2:
The patent varies key parameters such as training datasets, model architectures, and processing methods across different driving models. By changing these parameters, each model develops different characteristics and strengths, enabling the system to generalize better across diverse driving scenarios. This parameter variation allows the system to adapt to different user behaviors while maintaining a relatively simple overall structure.
3Measurement precision
If multiple driving models are used to improve identification accuracy, then the identification accuracy improves, but the model training complexity and time increase
Solution Approach 1:
The patent performs preliminary actions by pre-training multiple driving models on different datasets before actual use. The models are prepared in advance with their respective training completions, so that during actual identification operations, the system can quickly combine their outputs without requiring extensive real-time computation. This preliminary preparation reduces the time burden during deployment while maintaining high accuracy.
4Measurement precision
If multiple driving models are used to improve identification accuracy, then the identification accuracy improves, but the computational resources required increase
Solution Approach 1:
The patent segments the computational workload across multiple specialized models rather than using one comprehensive model. Each model is optimized for specific aspects of identification, allowing for more efficient computation in their respective domains. This segmentation enables the system to achieve high accuracy while managing computational resource consumption through targeted, specialized processing rather than brute-force comprehensive analysis.
Data Source
AI summary
A driving model training method, a driver identification method, an apparatus, a device and a medium are provided. The training method comprises: acquiring training behavior data of a user wherein the training behavior data are associated with a user identifier; acquiring training driving data associated with the user identifier based on the training behavior data; acquiring positive and negative samples from the training driving data based on the user identifier, and dividing the positive and negative samples into a training set and a test set; training the training set using a bagging algorithm, and acquiring an original driving model; testing the original driving model using the test set, and acquiring a target driving model. The training method effectively enhances generalization of the driving model, solves the problem of poor identification results of the existing driving identification model, and improves the accuracy rate of identifying driving of drivers.