This repository is to accompany the IJCAI-2019 publication:
"Twin-Systems to Explain Artificial Neural Networks using Case-Based Reasoning: Comparative Tests of Feature-Weighting Methods in ANN-CBR Twins for XAI"
Everything necessary to recreate the paper is included above in three folders with iPython notebooks.
-
Tabular with Sigmoid -- This is an example of twinning an MLP with a CBR system when using a Sigmoid output with tabular data in a binary class classification domain.
-
Tabular with SoftMax -- This is an example of twinning an MLP with a CBR system when using a SoftMax output with tabular data in a multi-class classification domain.
-
Deep Learning -- This is an example of twinning a CNN with a CBR system for image data in a classification domain.
Any questions or queries feel free to email me at [email protected] -- Thanks for reading!