Research on Improving the Trigger Success Rate of Automobile Cockpit Push Products Based on Data Mining
With the transformation of the automotive industry to intelligence and networking,the intelligent cockpit experience has become the core competitiveness of middle and high-end models.The traditional push function development method is limited by the randomness of manual configuration,resulting in low trigger success rate and poor touch accuracy.In order to solve these problems,an intelligent cockpit scene push model based on Internet of Vehicles data is proposed.Through in-depth analysis of the characteristics of functional data,the model uses Supervised Learning technology to train multiple algorithms,uses Semi-Supervised Learning method to overcome the challenge of insufficient data labels,and uses terminal data event tracking technology to finely optimize the model to improve the accuracy and efficiency of push.The experimental results show that compared with the traditional method,the model has achieved significant improvement in trigger success rate and touch accuracy.This achievement not only provides strong support for the development of intelligent cockpit technology but also brings more personalized and high-quality cockpit experience to users.
data miningInternet of Vehiclesvehicle intelligent cockpitsingle Decision Treegradient boosting Decision TreeRandom ForestSemi-Supervised Learningdata event tracking
国家科技期刊平台
NSTL
万方数据
