Handling Outliers in Clusters using Silhouette Analysis
Identify and remove outliers in each cluster from K-Means clustering
Published in
4 min readOct 20, 2020
The real-world data often has a lot of outlier values. The cause of outliers can be data corruption or failure to record data. The handling of outliers is very important during the data preprocessing pipeline as the presence of outliers can prevent the…