python-3.x - Does time series forecasting belong to supervised learning? or is it another category of machine learning?
问题描述
I have been analyzing several different methods of time series forecasting such as ARIMA and SARIMA using statsmodels library for my final year project. Checking on past literature I have seen that regression algorithms could also be used in combination with methods such as sliding window. However what I can't clarify is what type of algorithms does time series forecasting fall in to. I'm sure it is not unsupervised, therefore does this mean that time series forecasting algorithms are supervised algorithms? or is it a different type of machine learning?
解决方案
有很多技术可以解决时间序列预测问题,您可以使用监督或无监督学习,例如:使用无监督学习对时间序列进行聚类并使用聚类进行预测。最著名的技术之一是将数据集重新构建为监督学习数据集,以便您可以使用一些监督算法,当然这完全取决于您正在处理的用例,因此将时间序列算法限制为特定的我认为这种学习方式不是一个好主意。
推荐阅读
- next.js - 为什么我不能使用 QueryCache 从 queryClient.prefetchQuery 获取“缓存”
- git - 如何检查git下项目的所有变化?
- java - 我在一个布局中创建了两个回收器视图,并且我有 nullPointerException
- python - 当熊猫中的值为空时,to_list 不适用于熊猫
- sqlite - POST 找不到
- r - 为什么会出现“因子级别[2]重复”的错误?
- elk - ELK 设置的端口
- docker - 如何使用 prometheus 查找容器的状态?
- android-jetpack - Jetpack Compose - 如何向可组合添加多个修饰符,顺序是否重要?
- xamarin.forms - Visual Studio Xamarin 表单不显示设计视图