Consider using the frogs species dataset (frogs_species.csv) for this discussion question. For this dataset, please do the following:

Describe the dataset; its main features, labels, clusters, and usage.

Select all features in the dataset then apply a K-means clustering algorithm with a different number of clusters.

Evaluate the results of each run and create a chart that visualizes the results of each number of clusters.

From your point of view, what is the best number of clusters and why?

hint : python code # data mining

you can find the dataset on Kaggle