Research Topic: Interpretable Machine Learning

**ML-ModelExplorer: An explorative model-agnostic approach to evaluate and compare multi-class classifiers**

A major challenge during the development of Machine Learning systems is the large number of models resulting

from testing different model types, parameters, or feature subsets. The common approach of selecting the best model using one overall metric does not necessarily find the most suitable model for a given application, since it ignores the different effects of class confusions.

Expert knowledge is key to evaluate, understand and compare model candidates and hence to control the training process.

To address this problem, ML-ModelExplorer is proposed — an explorative, interactive, and model-agnostic approach utilising confusion matrices. It allows to conduct a thorough and efficient evaluation of multiple models by taking overall metrics, per-class errors, and individual class confusions into account.

**ML-ModelExplorer: An Explorative Model-Agnostic Approach to Evaluate and Compare Multi-class Classifiers.**

Theissler A., Vollert S., Benz P., Meerhoff L.A., Fernandes M. (2020). In: Machine Learning and Knowledge Extraction. Proceedings CD-MAKE 2020. Lecture Notes in Computer Science, vol 12279. pages 281-300. ISBN: 978-3-030-57320-1, Springer, Cham. **Download** , **Link to paper**

To import own data, follow this format:

CSV-file with concatenation of confusion matrices.

Example for 3 classes and 3 models (columns = class labels, rows = predictions):

class1,class2,class3

70,30,0

20,150,50

10,20,450 <— end of first confusion matrix

85,5,0

10,190,20

5,5,480

95,5,0

3,180,100

2,15,400