Home

delfos: DEveLopers' Framework of Suggestions is a recommender systems library that has a number of recommendation algorithms, datasets, experimentation protocols and metrics to perform experiments to evaluate and compare recommendation techniques.

Althoug delfos's main aim is to provide a simple library for researchers who propose new recommendation algorighms, it can also be used for educational and personalisation purposes.

Additionally, delfos provides a simple and powerfull framework to implement and integrate your recommendation algorithm.

Research

• Compare a personalisation algorithm with state-of-the-art approaches.
• Single-user recommendation experimentation.
• Group recommendation experimentation.
• Group-level recommendation experiments.

Learning

All the source code is available, which allows to explore the code and learn about the concrete implementations of the state-of-the-art methods.

Personalisation

Integrate personalisation techniques in any system.

Compare how differents systems perform on a given domain. This helps websites administrators to chose the best recommendation approach.

Adjust the recommender systems configuration to get the optimal performance.