又被Mining Massive Datasets的老师伤了!
Mining Massive Datasets这周的课讲聚类和计算广告学:二分图匹配。课后的作业好几个都是一眼看不出来,只好写程序算。
其中有一道题是这个样子: We wish to cluster the following set of points: into 10 clusters. We initially choose each of the green points(25,125), (44,105), (29,97), (35,63), (55,63), (42,57), (23,40),(64,37), (33,22), and (55,20) as a centroid. Assign each of thegold points to their nearest centroid. (Note: the scales of thehorizontal and vertical axes differ, so you really need to applythe formula for distance of points; you can’t just “eyeball” it.)Then, recompute the centroids of each of the clusters.