Low-code ML with KNIME: Building Clustering Models

Clustering is an unsupervised learning technique that finds logical groupings or clusters in your data for example identifying what social network users have the same interests and background. In this course explore how clustering models seek to find logical groupings in your data. Next construct a KNIME workflow to load and explore data for a clustering model. Then fill in missing values using different imputation techniques identify highly correlated variables and deal with outliers. Fit a k-means clustering model on your data identify clusters and use scatter plots to visualize the clusters in your data. Finally perform dimensionality reduction using principal component analysis (PCA) and use the silhouette score to evaluate the number of clusters that gives you the best clustering for your data. Upon course completion you will be able to fit and evaluate clustering models on your data and visualize clusters using 2-D and 3-D visualizations.