MongoDB query for ranking / search count?

To implement ranking or search count in MongoDB, use the aggregation pipeline with $setIntersection to find matching elements and $divide to calculate a relevance score based on the ratio of matched terms to total document elements.

Syntax

db.collection.aggregate([
    { "$match": { "fieldName": { "$in": searchArray } } },
    { "$addFields": {
        "rankScore": {
            "$divide": [
                { "$size": { "$setIntersection": ["$fieldName", searchArray] } },
                { "$size": "$fieldName" }
            ]
        }
    }},
    { "$sort": { "rankScore": -1 } }
])

Sample Data

db.demo120.insertMany([
    {
        "Name": "Chris",
        "Subjects": ["MySQL", "MongoDB", "Java", "Python"]
    },
    {
        "Name": "Bob", 
        "Subjects": ["C", "MongoDB"]
    }
]);

{
    "acknowledged": true,
    "insertedIds": [
        ObjectId("5e2f11aed8f64a552dae6365"),
        ObjectId("5e2f11afd8f64a552dae6366")
    ]
}

Display Sample Data

db.demo120.find();

{ "_id": ObjectId("5e2f11aed8f64a552dae6365"), "Name": "Chris", "Subjects": ["MySQL", "MongoDB", "Java", "Python"] }
{ "_id": ObjectId("5e2f11afd8f64a552dae6366"), "Name": "Bob", "Subjects": ["C", "MongoDB"] }

Example: Ranking Query

var searchTerms = ["MySQL", "Java", "MongoDB"];

db.demo120.aggregate([
    { "$match": { "Subjects": { "$in": searchTerms } } },
    {
        "$addFields": {
            "RankSearch": {
                "$divide": [
                    { "$size": { "$setIntersection": ["$Subjects", searchTerms] } },
                    { "$size": "$Subjects" }
                ]
            }
        }
    },
    { "$sort": { "RankSearch": -1 } }
]);

{ "_id": ObjectId("5e2f11aed8f64a552dae6365"), "Name": "Chris", "Subjects": ["MySQL", "MongoDB", "Java", "Python"], "RankSearch": 0.75 }
{ "_id": ObjectId("5e2f11afd8f64a552dae6366"), "Name": "Bob", "Subjects": ["C", "MongoDB"], "RankSearch": 0.5 }

How It Works

• $setIntersection finds common elements between document array and search terms
• $divide calculates relevance score: (matches / total_document_elements)
• $sort orders results by highest relevance score first
• Chris scores 0.75 (3 matches out of 4 subjects), Bob scores 0.5 (1 match out of 2 subjects)

Conclusion

MongoDB's aggregation pipeline enables sophisticated ranking by calculating relevance scores using $setIntersection and $divide. This approach ranks documents based on the percentage of matching elements, providing meaningful search results ordered by relevance.