MongoDB aggregation to combine or merge fields and then count?

To combine or merge fields and then perform count, use $group along with $sum and $sort. Let us create a collection with documents −

> db.demo647.insertOne({"Subject":"MySQL"});
{
   "acknowledged" : true,
   "insertedId" : ObjectId("5e9c86316c954c74be91e6ee")
}
> db.demo647.insertOne({"Subject":"MongoDB"});
{
   "acknowledged" : true,
   "insertedId" : ObjectId("5e9c86356c954c74be91e6ef")
}
> db.demo647.insertOne({"Subject":"MySQL"});
{
   "acknowledged" : true,
   "insertedId" : ObjectId("5e9c86376c954c74be91e6f0")
}
> db.demo647.insertOne({"Subject":"SQL Server"});
{
   "acknowledged" : true,
   "insertedId" : ObjectId("5e9c86406c954c74be91e6f1")
}
> db.demo647.insertOne({"Subject":"MongoDB"});
{
   "acknowledged" : true,
   "insertedId" : ObjectId("5e9c86436c954c74be91e6f2")
}
> db.demo647.insertOne({"Subject":"PL/SQL"});
{
   "acknowledged" : true,
   "insertedId" : ObjectId("5e9c864b6c954c74be91e6f3")
}
> db.demo647.insertOne({"Subject":"MongoDB"});
{
   "acknowledged" : true,
   "insertedId" : ObjectId("5e9c86c16c954c74be91e6f4")
}

Display all documents from a collection with the help of find() method −

> db.demo647.find();

This will produce the following output −

{ "_id" : ObjectId("5e9c86316c954c74be91e6ee"), "Subject" : "MySQL" }
{ "_id" : ObjectId("5e9c86356c954c74be91e6ef"), "Subject" : "MongoDB" }
{ "_id" : ObjectId("5e9c86376c954c74be91e6f0"), "Subject" : "MySQL" }
{ "_id" : ObjectId("5e9c86406c954c74be91e6f1"), "Subject" : "SQL Server" }
{ "_id" : ObjectId("5e9c86436c954c74be91e6f2"), "Subject" : "MongoDB" }
{ "_id" : ObjectId("5e9c864b6c954c74be91e6f3"), "Subject" : "PL/SQL" }
{ "_id" : ObjectId("5e9c86c16c954c74be91e6f4"), "Subject" : "MongoDB" }

Following is the query to combine or merge fields then count −

> db.demo647.aggregate([ { "$group": { "_id": "$Subject", "COUNT": { "$sum": 1 } } }, { "$sort": { "COUNT": -1 } }, { "$limit": 2 } ] );

This will produce the following output −

{ "_id" : "MongoDB", "COUNT" : 3 }
{ "_id" : "MySQL", "COUNT" : 2 }