README.md

Vague-Conditions-in-Object-Search-Seminar

Movies Context Based Recommendation demo for aggregated movies set from Kaggle and THMD using Recombee under the hood.

Technology Stack (only for learning purpose ):

• Docker-Compose
• Node.js
• Vue.js + Vuetify.js
• JWT for authentication
• Mongodb
• Neo4j
• Recombee (Recommender as a Service)
• TMDB (Movies additional data)
• Kaggle (Movies data set)

Docker Compose up and running

1. Create accounts by Recombee and TMDB
2. Create creds.env in recommender/

   RECOMBEE_DB=db_name
   RECOMBEE_TOKEN=token
   TMDB_API_KEY=api_key

3. Import movie data using the script (comming soon)

   cd recommeder
   node importMovies.js
   

4. Run the system

   docker-compose up -d 
   

5. Run the client (not dockerized yet )

   cd client
   npm run serve
   

6. Go to localhost:8080, create new user and enjoy

Notes

Neo4j UI on localhost:7474

URL: bolt://localhost/7687
USER: neo4j
PASSWORD: admin