User Segmentation API

Example of a simple implementation to build a user segmentation api on top of Redis. This can for instance be used to categorize users into segment for ads or recommendations.

The client should post the tags each consumed video. This backend will set the tags on the user and increment its weight if it already exists. over time, default 30 days, the weight will decrease. So the segmentation of interests should always be up to date with the users latests interests.

Related repositories

• Ratings API
• Favorites API
• Continue Watching API
• Stream Limit API
• Recommendation Engine

Requirements

• nodejs v10+
• redis

Usage

• git clone git@github.com:Eyevinn/user-segmentation-api.git
• cd user-segmentation-api
• npm install
• Start Redis locally or insert the needed keys into the .env file
• npm start to run the server

Endpoints

• POST /segmentation/:userId/:tags To set or update the weight of the tags on the user (tags comma separated)
• GET /segmentation/segment/:tag To get the number of users tagged with the specified tag.
• GET /user/:userId/segments To get a sorted list of the segments which the user belongs to. Weighted.
• GET /user/:userId/interestTo get a single interest randomly selected, but with weight in concern.

Examples are available in the example.http file.

Environment variables

• NODE_ENV if set to development there will be some logging made into the console
• REDIS_URL if not local
• REDIS_PORT if not default (6379)
• REDIS_AUTH
• SEGMENTATION_TTL for how long a segment input should be valid, before decreased in weight. (Default 30 days)

Docker

A docker-compose config file is provided that takes care of building the image and running this container together with a Redis db container.

Start the service:

• docker-compose up

Stop the service:

• docker-compose down

The Redis container is using /tmp as persistant storage but this can be changed by modifying the docker-compose.yml file. Change:

    volumes:
      - /tmp:/bitnami/redis/data

to where you want the persistant storage to be located.

About Eyevinn Technology

Eyevinn Technology is an independent consultant firm specialized in video and streaming. Independent in a way that we are not commercially tied to any platform or technology vendor.

At Eyevinn, every software developer consultant has a dedicated budget reserved for open source development and contribution to the open source community. This give us room for innovation, team building and personal competence development. And also gives us as a company a way to contribute back to the open source community.

Want to know more about Eyevinn and how it is to work here. Contact us at work@eyevinn.se!