Close

Cluster Centroids @ Netflix Data Mining Challenge

17770 Movies, 480189 Reviewers and over 100 million ratings to correlate; this is when your desktop starts looking like this.

Netflix's Challenge is a very interesting algorthmic problem but beware, it will eat up all your time. Mean, Median, Mode, Variance, Averages, Correlations, neighborhoods and what not, you'd start dreaming in those. The amoung of data is huge is so it makes an excellent case for distributed computing.

parallel-processing.jpg

References

Share