--

Very interesting,

This is similar to the LIME (Local Interpretable Model-Agnostic Explanations) technique used on NLP, which tries to figure out which word most impacts the tagging result.

--

--

Abdelkader Rhouati

Data scientist & Ph.D. researcher on AI. My area of expertise is around Deep Learning, NLP, and XAI — https://abdelkader-rhouati.medium.com/membership