首页
›
English Technical Courses
›
035-AI for CAE Simulation GNN and VVUQ
›
《Random Forest and XGBoost Occupant Injury Prediction 100 Cases》
VIP20
本课程属于：035-AI for CAE Simulation GNN and VVUQ
English Technical Courses · 共 20 门相关课程，组合学习更完整。
查看所属VIP全部课程 →
← 返回所属目录
咨询 / 购买
Random Forest and XGBoost Occupant Injury Prediction 100 Cases
100 Lessons
01
Load and Audit the Occupant Injury Dataset
02
Define Injury Severity Target and Leakage-Safe Feature Set
03
Stratified Train-Validation-Test Split for Imbalanced Injury Classes
04
Baseline Logistic Regression Benchmark for Injury Prediction
05
Random Forest Baseline with Default Hyperparameters
06
XGBoost Baseline with Default Hyperparameters
07
Handling Missing Occupant and Crash Variables
08
Encoding Categorical Crash and Vehicle Variables
09
Class Imbalance Handling with Class Weights
10
Threshold Tuning for Severe Injury Recall
11
Random Forest Hyperparameter Grid for Injury Prediction
12
XGBoost Hyperparameter Search with Early Stopping
13
Feature Importance Comparison Between Random Forest and XGBoost
14
Permutation Importance Validation of Top Injury Predictors
15
Partial Dependence of Injury Risk on Impact Speed
16
SHAP Analysis of Severe Injury Drivers
17
Cross-Validated Model Comparison with Repeated Stratified Folds
18
Calibration Curves for Injury Probability Outputs
19
Probability Calibration with Isotonic and Platt Scaling
20
Cost-Sensitive Evaluation of Injury Misclassification
21
Precision-Recall Curve Analysis for Rare Injury Classes
22
Confusion Matrix Deep Dive at Clinical Operating Threshold
23
Random Forest Out-of-Bag Error as Validation Proxy
24
XGBoost Learning Curve and Overfitting Diagnosis
25
Random Forest Learning Curve for Sample Size Planning
26
Detecting and Mitigating Overfitting in Random Forest
27
Regularization Tuning for XGBoost on Injury Data
28
Subsample and Column Sample Tuning for XGBoost Stability
29
Random Forest Proximity-Based Occupant Similarity
30
XGBoost Tree Dump Inspection for Injury Rules
31
Random Forest Decision Path Extraction for High-Risk Occupants
32
Handling Rare Crash Types with Stratified Sampling
33
Subgroup Performance Audit by Occupant Age Group
34
Subgroup Performance Audit by Restraint Use
35
Subgroup Performance Audit by Vehicle Body Type
36
Fairness-Aware Evaluation Across Occupant Demographics
37
Mitigating Subgroup Disparity with Sample Reweighting
38
Feature Interaction Discovery with XGBoost SHAP Interaction Values
39
Random Forest Feature Interaction via Conditional Inference
40
Recursive Feature Elimination for Injury Predictors
41
Boruta-Style Feature Selection for Random Forest
42
Dimensionality Reduction with PCA Before Ensemble Training
43
Handling High-Cardinality Crash Location Variables
44
Temporal Validation with Crash Year Holdout
45
Detecting Distribution Shift Between Training and Deployment Years
46
Domain Adaptation via Sample Weighting for New Crash Year
47
Ensembling Random Forest and XGBoost Predictions
48
Stacking with a Logistic Meta-Learner for Injury Risk
49
Blending Random Forest and XGBoost with Optimized Weights
50
Model Selection Decision Record for Injury Prediction
51
Serializing and Versioning Trained Injury Models
52
Building a Batch Scoring Pipeline for New Crash Records
53
Input Schema Validation for Scoring New Occupant Data
54
Monitoring Prediction Drift in Deployed Injury Model
55
Retraining Trigger and Rollback Strategy for Injury Models
56
Unit Tests for Injury Preprocessing and Feature Engineering
57
Regression Tests for Model Prediction Stability
58
Reproducibility Audit with Fixed Seeds and Environment Pinning
59
Hyperparameter Search Reproducibility and Search Space Documentation
60
Comparing Random Forest and XGBoost Training Time and Resource Use
61
Inference Latency Optimization for Real-Time Injury Scoring
62
Reducing Model Size for Edge Deployment in Vehicles
63
Cross-Validation Strategy Comparison for Small Injury Datasets
64
Bootstrapped Confidence Intervals for Injury Model Metrics
65
Statistical Significance Testing Between Model AUCs
66
Learning to Rank Occupants by Injury Risk
67
Quantile Regression with XGBoost for Injury Severity Scores
68
Multi-Class Injury Severity Prediction with XGBoost
69
Ordinal Injury Severity Modeling with Random Forest
70
Handling Label Noise in Injury Severity Records
71
Semi-Supervised Learning with Unlabeled Crash Records
72
Active Learning for Efficient Injury Label Acquisition
73
Data Augmentation for Rare Severe-Injury Cases
74
Focal Loss Style Weighting for XGBoost Injury Classification
75
Random Forest with Balanced Bootstrap Sampling
76
Outlier Detection in Occupant Injury Features
77
Robust Loss Training for XGBoost Under Outliers
78
Feature Scaling Sensitivity Analysis for Tree Ensembles
79
Monotonic Constraint Enforcement in XGBoost Injury Models
80
Interaction Constraint Testing for Physically Plausible Injury Risk
81
Counterfactual Explanation for Individual Injury Predictions
82
Local Interpretable Surrogate for Ensemble Injury Predictions
83
Anchoring Explanations for Severe Injury Predictions
84
Global Surrogate Decision Tree for Ensemble Behavior
85
Feature Attribution Consistency Across Random Forest and XGBoost
86
Model Disagreement Analysis for High-Risk Occupants
87
Error Analysis of False Negatives in Severe Injury Prediction
88
Error Analysis of False Positives in Severe Injury Prediction
89
Residual Analysis for Injury Severity Regression Variant
90
Handling Censored or Truncated Injury Severity Records
91
Imputation Model Comparison for Missing Injury Severity Labels
92
Sensitivity Analysis to Imputation Strategy Choice
93
Feature Stability Selection Across Bootstrap Resamples
94
Model-Based Occupant Risk Stratification for Policy Analysis
95
What-If Scenario Analysis for Restraint System Design
96
Impact Speed Threshold Estimation from Ensemble Predictions
97
Occupant Age and Injury Risk Interaction Profiling
98
Restraint Effectiveness Estimation with Ensemble Models
99
Vehicle Body Type Risk Ranking from Model Predictions
100
Crash Type Severity Profiling with Partial Dependence
所属VIP或分类
同VIP：Active-Learning-and-Adaptive-LS-DYN
同VIP：AI-Model-Verification-Validation-an
同VIP：AI-Prediction-of-Airbag-Deployment-