基于LightGBM的智能可穿戴设备用户行为预测

Predicting User Behavior Based on LightGBM for Smart Wearable Devices

智能可穿戴设备产生的大量数据是人类宝贵的数字资源。使用开放数据集和主流数据分析工具,如可进行快速模型开发的PyCaret模块,有助于人们进行数据挖掘工作,且不被细节所困扰。作为Kaggle竞赛爱好者的常用工具,LightGBM分类器对用户行为的预测表现优异,对此文中的研究结果也得到验证。

The large volume of data generated by smart wearable devices is a valuable digital resource for humanity.Using open datasets and mainstream data analysis tools,such as the PyCaret module for rapid model development,helps people conduct data mining without being bogged down by details.As a commonly used tool among Kaggle competition enthusiasts,LightGBM demonstrates excellent performance in predicting user behavior,and this has been validated by the research results presented in this paper.