A recommendation engine in your PHP applications
summary
An online bookshop prototype uses Prediction.io with an events server and engines (Universal Recommender template) to record view/like/add to basket/purchase actions, train nightly, and serve both “who bought this also bought” item similarities and per-user suggestions. Deployment uses an Amazon AMI or HBase/Elasticsearch (or MySQL/Postgres for smaller sites). Four engines are created initially to work around limitations; recommendations go live on product pages and power personalized emails and the client reports satisfaction.
AI-written summary · about
chapters (8)
- 0:00Start
- 4:38Content based
- 5:51Collaborative filtering
- 12:08Choosing the tool
- 14:10The event server
- 20:11Engines
- 22:29Implementation
- 25:35Alternative approaches
related talks
39mA recommendation engine for your applications
46mYour own recomendation Engine with Neo4j AND Reco4PHP
55mBuilding a Recommendation Engine with Machine Learning Techniques
31mCreating recommender systems with Spark, Scala and Prediction.Io
32mRecommender Engines - What are they? Where do they live? What do they feed on?
28mThe Power of Personalized Customer Experiences Through AI Innovations
49mAn Introduction to Recommendation Engines
29mThe Challenges of Building a Personalized Recommendation System in E-Commerce