Discovery is a key challenge within Information Management. While many approaches focus on metadata or full text search, the recent trend has seen a rise in the role that social influences can offer.
By combining Artificial Intelligence—specifically Machine Learning collaborative filtering techniques - with a data set of user preferences, we can produce personalised content recommendations.
Apache Mahout is an Open Source scalable machine learning library that can be integrated with the Alfresco Rating Service.
This talk will:
- Give an introduction to collaborative filtering algorithms
- Show you how to integrate 5 star ratings into the Share UI
- Detail the Mahout integration points and options
- Teach you how to evaluate recommender settings
- Demonstrate the end–to-end flow of preferences through to recommendations
This will be a technical session for developers and architects and will include examples, code and a demonstration.
The session recording is from the version of this talk given in London.