基于睡眠传感器的身体疲劳等级预测数据

基于睡眠传感器采集的用户睡眠心率数据，为用户的身体疲劳程度评价提供数据支持。此数据主要用于用户的健康管理服务，将身体疲劳程度按从低到高顺序分为5个等级，当疲劳等级较高时，及时提醒用户注意休息，并多留意自身的健康变化情况。1、 预处理夜间心率序列数据，对心率序列进行重采样，并填充缺失值;
2、 对平均心率序列进行三次样条插值拟合，得到平均心率序列的整体变化趋势；
3、 定义5种平均心率变化趋势典型模式，该5种模式从用户反馈、医学理论和大数据统计分析三方面总结归纳而来，分别对应了身体疲劳程度等级1、2、3、4、5；
4、 将当前平均心率趋势序列与上述5种模式分别进行相似性计算，选取计算结果中相似性程度最高的一项，作为当前平均心率变化趋势模式；
5、 根据平均心率变化模式与身体疲劳等级之间的对应关系，得到当前的身体疲劳等级。

This dataset consists of user sleep heart rate data collected via sleep sensors, which provides data support for evaluating users' physical fatigue levels. The data is primarily used for user health management services. Physical fatigue levels are divided into 5 grades from low to high. When the fatigue grade is relatively high, users will be promptly reminded to pay attention to rest and closely monitor their own health changes.
1. Preprocess the nighttime heart rate sequence data: resample the heart rate sequence and impute missing values;
2. Perform cubic spline interpolation fitting on the average heart rate sequence to obtain the overall change trend of the average heart rate sequence;
3. Define 5 typical patterns of average heart rate change trends. These 5 patterns are summarized from three aspects: user feedback, medical theory and big data statistical analysis, and correspond to physical fatigue grades 1, 2, 3, 4 and 5 respectively;
4. Calculate the similarity between the current average heart rate trend sequence and each of the above 5 patterns, and select the one with the highest similarity as the current average heart rate change trend pattern;
5. Obtain the current physical fatigue grade based on the corresponding relationship between the average heart rate change pattern and the physical fatigue grade.