Problem Description

In this assignment, you will apply clustering algorithms to three toy datasets. You can try whatever clustering algorithm with whatever parameter setting. In the end, you need to report the "best" clustering output with a visualized $2$D plot.

You can use a third$-$party library's implementation of clustering algorithms.

Your solution should include at least $3$ different clustering algorithms $($which could be the ones not covered in lectures$)$ for comparison.

You need to pick one validation metric $($e$.$g$.,$ SSE$)$ to justify the quality of the clustering output.

You should document the parameter tuning steps $($even if it is done manually$).$

Input Files

We provide $3$ toy datasets $($D$01.$csv$,$ Download D$01.$csv$,$D$02.$csv$,$ Download D$02.$csv$,$D$03.$csv Download D$03.$csv$)$ for this assignment, which are all $2$D datasets in the format of

for each line. You are encouraged to have them plot out in the first place to give you some hints on what clustering algorithms would fit.

Key Outputs

Your program should at least plot the best clustering results and specifically indicate how many clusters are found for a given dataset.

Assessment

Your program should be compilable and executable. Clear documentation on any preprocessing, parameter tuning, and validation steps.