mapreduce - coding for multiple key and values - hadoop

Question

In need to emit two keys and two values from my mapper. could you please provide me info , how to write code and data type for that. for example :-
key = { store_id : this.store_id,
product_id : this.product_id };
value = { quantity : this.quantity,
price : this.price,
count : this.count };
emit(key, value);
regards

Answer 1

As per the given example, A B B C A R A D S D A C A R S D F A B
From the mapper emit
key - A
value A, AB
key - B
value B,BB
key - B
value B, BC
key - C
value C, CA
and so on...
In the reducer, you get the grouped values
key - A
values A, AB, A, AR, A, AD, A, AC and so on
key - B
value - B, BB,B,BC and so on
Add a delimiter of your choice between the 2 words/alphabets
for each key in reducer, you can use a hashmap/mapwritable to track the occurrence count of each value
ie for example
A - 5 times
AB - 7 times
etc etc
Then you can calculate the ratio
Sample Mapper Implementation
public class TestMapper extends Mapper<LongWritable, Text, Text, Text> {
#Override
public void map(LongWritable key, Text value, Context context)
throws IOException, InterruptedException {
String[] valueSplits = value.toString().split(" ");
for(int i=0;i<valueSplits.length;i++){
if(i!=valueSplits.length-1){
context.write(new Text(valueSplits[i]),new Text(valueSplits[i]+"~"+valueSplits[i+1]));
}
context.write(new Text(valueSplits[i]), new Text(valueSplits[i]));
}
}
}
Sample Reducer Implementation
public class TestReducer extends Reducer<Text, Text, Text, Text> {
public void reduce(Text key, Iterable<Text> values, Context context)
throws IOException, InterruptedException {
Map<String,Integer> countMap= new HashMap<String,Integer>();
for(Text t : values){
String value = t.toString();
int count =0;
if(countMap.containsKey(value)){
count = countMap.get(value);
count+=1;
}else{
count =1;
}
countMap.put(value, count);
}
for(String s : countMap.keySet()){
if(s.equalsIgnoreCase(key.toString())){
}else{
int keyCount = countMap.get(s.split("~")[0]);
int occurrence = countMap.get(s);
context.write(new Text(key.toString()+" , "+s), new Text(String.valueOf((float)occurrence/(float)keyCount)));
}
}
}
}
For an input of
A A A B
the reducer would emit
A , A~A 0.6666667
A , A~B 0.33333334
AA appears 2 times, AB 1 time and A 3 times.
AA is hence 2/3
AB is hence 1/3