Unsupervised Machine Learning Approaches for Outlier Detection in Time Series, using Python
Published in
8 min readAug 23, 2019
In this post, I cover some of my favorite methods for detecting outliers in time series data. There are many different approaches for detecting anomalous data points; for the sake of brevity, I only focus on unsupervised machine learning methods in this post.
The anomaly/outlier detection algorithms covered in this article include:
- Low-pass filters: taking the centered rolling average of a time series, and removing anomalies based on Z-score