Implementation of Anomaly Detection in weather data using unsupervised Isolation Forest over a timeline.
Isolation forest is a machine learning algorithm for anomaly detection. It is an unsupervised learning algorithm that identifies anomaly by isolating outliers in the data. This concept is used to find anomalies in the weather data over a time period. Given below are some images of the anomalies detected.
- Humidity
- Pressure
- Temperature
A clear explanation of the mathematics behind this concept is given in the report.
The below table gives the results of Isolation forest for the New York city weather data.
This is a part of the mini project for Machine Learning Special topic course.
For doubts and contributions pleas contact [email protected]