e1 e2

pig-twt-e1.pig

+dataset = LOAD '/laboratory/twitter-small.txt' AS (id: long, fr: long);
+
+-- TODO: check if user IDs are valid (e.g. not null) and clean the dataset
+B = FILTER dataset BY (id>0);
+
+-- TODO: organize data such that each node ID is associated to a list of neighbors
+C = FOREACH B GENERATE fr;
+
+D = GROUP C BY fr;
+
+friends = FOREACH D GENERATE group, COUNT(C);
+
+STORE friends INTO '/twitter/results-e1';

pig-twt-e1.pig~

+dataset = LOAD '/laboratory/twitter-small.txt' AS (id: long, fr: long);
+
+-- TODO: check if user IDs are valid (e.g. not null) and clean the dataset
+B = FILTER dataset BY (id>0);
+
+-- TODO: organize data such that each node ID is associated to a list of neighbors
+C = FOREACH B GENERATE fr;
+
+D = GROUP C BY fr;
+
+friends = FOREACH D GENERATE group, COUNT(C);
+
+STORE friends INTO '/twitter/results1';

pig-twt-e2.pig

+dataset = LOAD '/laboratory/twitter-small.txt' AS (id: long, fr: long);
+
+-- TODO: check if user IDs are valid (e.g. not null) and clean the dataset
+B = FILTER dataset BY (id>0);
+
+-- TODO: organize data such that each node ID is associated to a list of neighbors
+C = FOREACH B GENERATE id;
+
+D = GROUP C BY id;
+
+friends = FOREACH D GENERATE group, COUNT(C) AS count;
+
+friends2 = FILTER friends BY (count > 2)
+
+STORE friends2 INTO '/twitter/results-e2';

pig-twt-e2.pig~

+dataset = LOAD '/laboratory/twitter-small.txt' AS (id: long, fr: long);
+
+-- TODO: check if user IDs are valid (e.g. not null) and clean the dataset
+B = FILTER dataset BY (id>0);
+
+-- TODO: organize data such that each node ID is associated to a list of neighbors
+C = FOREACH B GENERATE id;
+
+D = GROUP C BY id;
+
+friends = FOREACH D GENERATE group, COUNT(C) AS count;
+
+friends2 = FILTER friends BY (count > 2)
+
+STORE friends2 INTO '/twitter/results1';

pig-twt.pig

@@ -10,4 +10,4 @@ D = GROUP C BY id;
 
 friends = FOREACH D GENERATE group, COUNT(C);
 
-STORE friends INTO '/twitter/results';
+STORE friends INTO '/twitter/results1';

pig-twt.pig~

-dataset = LOAD 'tw.txt' AS (id: long, fr: long);
+dataset = LOAD '/laboratory/twitter-small.txt' AS (id: long, fr: long);
 
 -- TODO: check if user IDs are valid (e.g. not null) and clean the dataset
 B = FILTER dataset BY (id>0);
@@ -10,4 +10,4 @@ D = GROUP C BY id;
 
 friends = FOREACH D GENERATE group, COUNT(C);
 
-STORE friends INTO './local-output/OSN/twc/';
+STORE friends INTO '/twitter/results';