Using KMeans unsupervised classification algorithm to identify optimal number of clusters to divide a dataset
In this notebook, we use a randomly generated dataset from make_blobs library. We use the KMeans clustering algorithm to create a model and fit our dataset to it to help determine its class. Using elbow method, We then identify the ideal number of clusters by plotting the SSE (sum of squeared erros) for each cluster. The ideal number of cluster will be at the elbow of the plot , any additional clusters after which will be redundant.
