Two data files (a and b) are given below. (2-dimensional data) They are to be clustered by k-means, using the centers of the three circular regions as initial seeds. (Why did I say to use the center of each circle as initial seed, since k-means would not be so clever as to do so? I did because in part (b) the goal is to show that even if we tell k-means the right answer, k-means will insist on giving us back the wrong answer.) Use a k-means implementation (either an existing one or an implementation of your own) and perform it on the two datasets and show the final clustering results, using the provided right centroids (see the attached data files) as the initial seeds (k=3). Briefly explain the result of (b).

2A Dataset:

797	Centroid 1= (2.8, 5) Centroid 2=(7.7, 1) Centroid 3=(7.8, 9)
X	Y
0.709093	5.42129
0.750159	5.52021
0.709221	5.37187
0.688731	5.30593
0.731419	4.77882
0.688816	5.27299
0.729882	5.3719
0.729882	5.3719
0.729711	5.4378
0.729669	5.45427
0.750287	5.47078
0.750244	5.48726
0.750244	5.48726
0.750202	5.50373
1.32602	6.54257
0.770607	5.60261
0.770564	5.61909
0.791182	5.63559
0.791182	5.63559

2B dataset:

774
X	Y
1.03406	5.62529
0.986664	5.62539
0.98672	5.64631
1.01092	5.83454
1.01053	5.6881
1.01075	5.77178

Note: posting only few data points because of space issue while posting question.

please let me know how to upload two excel sheets here, or give me an email id so that i can send you the files