There is big sparsity and points separation or separability is hard in your plotting. I may suggest you MATLAB [ https://au.mathworks.com/help/stats/classification-using-nearest-neighbors.html ]

https://au.mathworks.com/help/stats/k-means-clustering.html?searchHighlight=k%20means%20clustering&s_tid=srchtitle_k%20means%20clustering_1

PYTHON ==== [https://www.askpython.com/python/examples/plot-k-means-clusters-python ]

Dear Jennifer Leong ,

Clustering is one of the most common exploratory data analysis technique used to get an intuition about the structure of the data. It can be defined as the task of identifying subgroups in the data such that data points in the same subgroup (cluster) are very similar while data points in different clusters are very different. In other words, we try to find homogeneous subgroups within the data such that data points in each cluster are as similar as possible according to a similarity measure such as euclidean-based distance or correlation-based distance. The decision of which similarity measure to use is application-specific.

https://towardsdatascience.com/k-means-clustering-algorithm-applications-evaluation-methods-and-drawbacks-aa03e644b48a

Regards,

Shafagat

First of all you need to refer back to your research objective(s). Because at most point in time, you may not necessarily have distinct clusters without little overlap. If your result of cluster analysis has explained your target objective then its OK.

The clusters may not be in perfect spherical shape, as cluster separation can be observed in your plot.

Best regds.

Isa Koki.

Check https://jakevdp.github.io/PythonDataScienceHandbook/05.11-k-means.html

https://www.askpython.com/python/examples/plot-k-means-clusters-python

A lower within-cluster variation is an indicator of a good compactness (i.e., a good clustering). The different indices for evaluating the compactness of clusters are base on distance measures such as the cluster-wise within average/median distances between observations. https://www.datanovia.com/en/lessons/cluster-validation-statistics-must-know-methods/

The plot is okay but you need to find the coefficient of correlation.

Dear Jennifer Leong ,

K-means algorithm depends on the distance function, (generally, it's squared euclidean distance), in this case clusters to be found have a spheric-ish shape. Using another distance function may affect on the results. Different features categories, plotting more than 2-dimensional data on 2D space, and the scale of plotting may also have a bad effect on results.

If your data is 2D, you can verify your result on matlab by the two built-in functions: k-means() and plot().

Kind regards,

Ibrahim.