TO display count of every hastag only in tweet info associated with given keyword.

Mapper:- this is the initial phase in the execution. In which every data node will be accessing data and every node will be having computation logic which was given by main.
These data nodes process the data, format or split based on key,value pairs.What will be key or value will be decided by main or can also be decided by users.
Reducer :- This is the last step in which key values obtained from mapper are combined here. The combination is also done in such a way that for each key all values in iterator mode are added and returned.

We all know a famous problem that can be solved using map reduce is word count problem. Here in a file or folder word with their frequencies are given as output that are occurring entirely all over.Aim of this part is to calculate total rows in a dataset.So the usual wordcount program can be developed to satisfy our aim . Instead of words we should make the mapper to take the whole json object or tweet row as a key. And the value for every json object associated with the key is 1. Because these objects are tweets and in a twitter data there cannot be repeated tweets. As tweets are non repeating, we can fix only value 1 with every tweet in mapper.
Now in reducer, all these 1s are added and returns total count. These total count ensures total tweets in a folder as we are adding all 1s associated as values with key rows.

1. First, set up your hadoop.
   
2. Next in partioner code change according to your keywords in quotes.
   
3. run commands according to your compiling environment.
   

Depending on number of keywords changed in the code that many partioned files will be created in which count of hashtags are generated.