rec-engine

Recommendation Engine

Overview

A Simple Colleabartive filtering recommendation engine written in Go using Redis as the In memory persistence store. The collabrative filtering algortithm was heavilly borrowed from

Swagger doc

Swagger documentation can be accessed here: http://localhost:3000/swagger/index.html

Usage

1. Run redis locally: docker run -p 6379:6379 --network redis-network --name some-redis -d redis

2. Install httpie or simply use curl

3. See samples below for all the API calls

======== CREATING SOME RATINGS ==========

 👉  http post localhost:3000/api/rate user=user2 item=item1 score:=5.6
HTTP/1.1 201 Created
Content-Length: 43
Content-Type: application/json
Date: Tue, 02 Mar 2021 08:44:49 GMT

{
    "Item": "item1",
    "Score": 5.6,
    "user": "user2"
}


 👉  http post localhost:3000/api/rate user=user2 item=item2 score:=5.6
HTTP/1.1 201 Created
Content-Length: 43
Content-Type: application/json
Date: Tue, 02 Mar 2021 08:44:59 GMT

{
    "Item": "item2",
    "Score": 5.6,
    "user": "user2"
}


👉  http post localhost:3000/api/rate user=user2 item=item3 score:=5.6
HTTP/1.1 201 Created
Content-Length: 43
Content-Type: application/json
Date: Tue, 02 Mar 2021 08:45:03 GMT

{
    "Item": "item3",
    "Score": 5.6,
    "user": "user2"
}


👉  http post localhost:3000/api/rate user=user2 item=item4 score:=8.6
^[[DHTTP/1.1 201 Created
Content-Length: 43
Content-Type: application/json
Date: Tue, 02 Mar 2021 08:45:10 GMT

{
    "Item": "item4",
    "Score": 8.6,
    "user": "user2"
}


👉  http post localhost:3000/api/rate user=user1 item=item1 score:=5.6
HTTP/1.1 201 Created
Content-Length: 43
Content-Type: application/json
Date: Tue, 02 Mar 2021 08:45:21 GMT

{
    "Item": "item1",
    "Score": 5.6,
    "user": "user1"
}

====== QUERYING SUGGESTIONS FOR A GIVEN USER =========

👉  http get localhost:3000/api/suggestion/user1                      
HTTP/1.1 200 OK
Content-Length: 88
Content-Type: application/json
Date: Tue, 02 Mar 2021 08:45:31 GMT

[
    {
        "item": "item4",
        "score": 8.6
    },
    {
        "item": "item3",
        "score": 5.6
    },
    {
        "item": "item2",
        "score": 5.6
    }
]

=========  Getting probability for a given user with an item ==========

 👉  http get localhost:3000/api/probability/user1/item2
HTTP/1.1 200 OK
Content-Length: 49
Content-Type: application/json
Date: Tue, 02 Mar 2021 08:49:12 GMT

{
    "item": "item2",
    "propability": 5.6,
    "user": "user1"
}

===== Get all the popular items ==============

👉  http get 34.89.236.249:9000/api/items
HTTP/1.1 200 OK
Content-Length: 58
Content-Type: application/json
Date: Mon, 08 Mar 2021 19:13:57 GMT

{
    "data": [
        {
            "item": "A3",
            "score": 2
        },
        {
            "item": "A2",
            "score": 1
        }
    ]
}

== Get all the items that the user rated / bought =======

👉  http get localhost:3000/api/items/user/A                  
HTTP/1.1 200 OK
Content-Length: 45
Content-Type: application/json
Date: Mon, 08 Mar 2021 19:19:02 GMT

{
    "data": [
        {
            "item": "A3",
            "score": 0
        }
    ],
    "user": "A"
}