I'm working on an unsupervised anomaly detection task on time series using isolation forest algorithm.
I'm developing in Python, more in detail using sklearn.
I found out a lot of examples on this, but what is not very clear, is how to set the contamination param during the instantion of IsolationForest.
Looking the documentation, contamination is
the amount of contamination of the data set, i.e. the proportion of outliers in the data set.
Should I use some statistical techniques to identify this percentage?