Data Science and Mining team (DaSciM)
LIX Laboratory, École Polytechnique.


MEKA: A Multi-label/multi-target Extension to weKA

I am the creator and main developer of the MEKA Project for multi-label classification: A Java framework containing around two dozen multi-label and multi-output classifiers for train/test, cross validation, incremental and semi-supervised learning.

Molearn: A Multi-label/multi-target framework in Python

Molearn provides a Python implementation of some of the classifiers included in MEKA (such as the meta labelling and probabilistic classifier chains frameworks), using numpy and sklearn. N.B. this project will soon be merged into the Scikit-MultiFlow project.

MOA: Massive Online Analysis

I am involved in the MOA Project for learning in data streams.

ALife: Artificial Life environment

ALife provides an environment with old RTS-style graphics, and several sample agents, for testing reinforcement learning algorithms.

A2RMS: Implementation of the A2RMS Sampling Algorithm

I (with Luca Martino and David Luengo) maintain Matlab/GNU Octave code for the A2RMS sampling algorithm.

DPFlib: A library in C for distributed particle filters.

I developed code for implementing distributed particle filters on very low powered sensor motes, in the TinyOS operating system. However, the code is also standalone and can be compiled and run by regular C. Here is a video of a deployment.


Datasets which I have created (or parsed into a new format) are available at; some of which are now also available from the mulan website.