Multiple Instance Learning Toolbox

Current version is 1.2.1 (Sept 2016).

This toolbox is an add-on to the PRTools toolbox. This version is adapted to work with Prtools 5. The toolbox contains algorithms to train, investigate, visualize and evaluate multiple instance learning classifiers. Realize that this toolbox is very preliminary still, because I am still not sure if the structure and the implementation choices that I made are sensible. But the code works:-)

If you want to use it, feel free. When you are putting these results in a publication, I would appreciate it when you would refer to this toolbox. For instance using this bibliography entry:

@misc{MIL2015,
	Author = {Tax, D.M.J., Cheplygina, V.},
	Month = {Jun},
	Note = {version 1.2.1},
	Title = {{MIL}, A {M}atlab Toolbox for Multiple Instance Learning},
	url = {http://prlab.tudelft.nl/david-tax/mil.html},
	Year = {2016}}

The toolbox is an extension of the PRTools toolbox, in which Matlab objects for prmapping and prdataset are defined. The data description toolbox uses these objects and their methods, but extends (and sometimes restricts) it to one-class classification. This means, that before you can use it to its full potential, you need to know a bit about PRTools. When you are completely new to pattern recognition, Matlab or PRTools, please familiarize yourself a bit with them first.

Download