Crab is a flexible, fast recommender engine for Python that integrates classic information filtering recommendation algorithms in the world of scientific Python packages (NumPy,SciPy, Matplotlib).

This poster will introduce the basic concepts about recommender systems and how you can build, use and evaluate your custom recommender systems with offline data and real word data using the Crab framework as also to invite more maintainers to help to improve and maintain the project.

This framework started around 2010 and it is an alternative for the Mahout, which is a popular one for machine learning written in Java. Several demonstrations will be presented during the talk to illustrate his work and how you can apply it on your web sites, e-commerces and real systems.

This project was presented at the Scipy Conference 2010 and the Brazilian Python Meeting 2010. For further information please visit the project home link:

https://github.com/muricoca/crab