The invention provides a model training method and a related device, a processing device can first acquire historical driving data and vehicle accident information corresponding to a plurality of vehicles in a preset historical period, and the vehicle accident information is used for identifying whether the plurality of vehicles are in an accident in the preset historical period. The processing device may determine a plurality of driving features respectively corresponding to the plurality of vehicles based on the historical driving data. The processing device can determine the associated parameters corresponding to the target driving features, and select the effective driving features having reasonable association with risk analysis from the plurality of driving features based on the associated parameters, so that an accident prediction model can be obtained through training based on the features, and the risk prediction accuracy is improved. Through the accident prediction model, the accident probability of the vehicle can be accurately predicted, so that the risk probability of the driver can be reduced from the dimension of analyzing the driving behavior of the driver, and the driving safety is improved.
本申请提供了一种模型训练方法和相关装置,处理设备可以先获取多个车辆在预设历史时段内分别对应的历史驾驶数据和车辆出险信息,该车辆出险信息用于标识多个车辆在所述预设历史时段内是否出险。处理设备可以基于该历史驾驶数据,确定多个车辆分别对应的多个驾驶特征。处理设备可以确定所述目标驾驶特征对应的关联参数,基于关联参数,从多个驾驶特征中选择对风险分析有合理关联性的有效驾驶特征,从而,可以基于这些特征,训练得到出险预测模型,通过该出险预测模型可以较为准确地预测车辆的出险概率,进而可以从对驾驶者驾驶行为分析的维度降低驾驶者的风险概率,提高行车安全性。
Model training method and related device
一种模型训练方法和相关装置
23.12.2022
Patent
Elektronische Ressource
Chinesisch
Europäisches Patentamt | 2022
|Traffic track prediction model training method, traffic track prediction method and related device
Europäisches Patentamt | 2024
|Europäisches Patentamt | 2024
|Battery detection method, prediction model training method and related equipment
Europäisches Patentamt | 2024
|