From the course: Python in Excel: Data Outputs in Custom Data Visualizations and Algorithms

Unlock this course with a free trial

Join today to access over 25,400 courses taught by industry experts.

Determining anomalies with anomaly detection algorithms

Determining anomalies with anomaly detection algorithms

From the course: Python in Excel: Data Outputs in Custom Data Visualizations and Algorithms

Start my 1-month free trial Buy for my team

Determining anomalies with anomaly detection algorithms

“

- [Presenter] On a high level, we can define an outlier as a deviation from where we are, while an anomaly is a deviation from where we expect to be. If we line up all the daily temperatures within a year for any given location, we would see a distribution where a lot of the data points are in the middle, but there are some at the ends. These outliers represent very hot or very cold days within the year, for example. However, within each season, sometimes we experience hotter or colder than expected average temperatures for that particular time of year. For example, a temperature of 80 degrees Fahrenheit might be hot for New York in the middle of winter, so it's an anomaly, but then in the summer, it might be an expected temperature that we see, so it's not. These outliers represent anomalies as temperatures. Let's check out how to run anomaly detection algorithms using Python code and Codespaces first. We'll use the DBSCAN section of the SK-Learn library to run the anomaly detection…

Contents