Our goal was to find the age group of users with the worst taste in movies. 
We defined this as being the group with the largest delta from the average 
for each movie. We used the MovieLens data

0.28103770286333096      for 25-34 with 35444.0 members
0.3431092428664245       for 18-24 with 24060.0 members
0.3538714921083055       for 35-44 with 19591.0 members
0.43826352357785453      for 50-55 with 6534.0 members
0.48737709237214044      for 56+ with 4990.0 members
0.5162371242145326       for 45-49 with 6890.0 members
0.6451387761243556       for Under 18 with 2491.0 members 

Based on this the Under 18 group was the group with the worst taste in movies.
		- run .jar in hadoop
    hadoop jar miner.jar smallTest smallOut
    - smallTest must contain your *.data files
    - smallOut must not exist
    - Your root directory in hadoop must contain at least one .user file
Currently input is expected in pipe delimited format, as found in the 
100,000 movieLens data

We realized after implementing our solution fully, that our metric 
was slanted towards simply giving the smallest group the label of 
worst taste. Since values from this group didn't play into the 
overall average as much, they tended to be near the top of the list
for poor taste. Given more time we would like to correct this issue
by adding in a factor for the group size.

MovieMiner.java - The main class
GetInterAverage.java - 1st M&R phase that gets the intermediate averages
ConsolidateInterAvg.java - 2nd M&R puts the individual group ratings 
                           into the 7 age groups and outputs
TaggedTuple.java - Supporting class
README - This file
miner.java - The compiled project
SmallTestFiles - The folder of files that we used for testing