Idea.txt

Dataset: Yelp dataset

Goal: Cluster restaurants (or some business) based on their location and rating. Afterwards, visualize the data.

Idea: Use the BFR clustering method in combination with MapReduce.

This will work in the same way as the counting itemsets work with MapReduce.
Step1: divide the data and let each worker cluster their data.
Step2: communicate the intermediate results to all other workers and recalculate based on these results.
Step3: Combine the information
Step4: visualize the data using for example Google map API.

I think it is even possible to do this maybe in one step: Map is using the bfr and Reduce will just merge the different clusters. 
If that is possible, I think it will be quite easy to implement.

==================================================================

Writing report: Rob (and Milan)
Visualization: Milan and Rob
Presentation: Milan and Rob
Setting up Hadoop with Yelp data: ?
Writing BFR: ?
Writing MapReduce to work with BFR: ?