Measuring model performance
There are a number of ways to measure the performance of a clustering model. Measuring the performance is important because it tells you how well each algorithm is doing in creating segments that have similar observations within each segment while appropriately differentiating segments from one another.
In this section, we’ll review three common performance measures: the Calinski-Harabasz score, the Davies-Bouldin score, and the Silhouette score. Diving into each of these measures further is outside the scope of this book. To continue your learning, we recommend checking out the following resources mentioned in each of the points, as follows:
- Calinski-Harabasz score: Higher scores indicate better models (https://scikit-learn.org/stable/modules/generated/sklearn.metrics.calinski_harabasz_score.html#sklearn.metrics.calinski_harabasz_score)
- Davies-Bouldin score: The minimum score is 0 and lower scores indicate better performance (https...
