Automatic fault type classification is an important ingredient of smart electrical grids. Similar to other machine-learning models, methods developed for fault classification suffer from the issue of lack of transparency. This work sheds light on preliminary insights of an ongoing study, in which we show how feature importance measurement and feature interaction visualization using partial dependence plots (PDPs) can help interpretability of the classification outcomes. While the former, measures the role of each feature on the final predictions in isolation, the latter focuses on mutual interaction between pairs of features. We show the merits of these two complementary feature analysis mechanisms in facilitating interpretability of the fault type classification task.

Interacting with features: Visual inspection of black-box fault type classification systems in electrical grids

Carmelo Ardito;Yashar Deldjoo;Eugenio Di Sciascio;Fatemeh Nazary
2020-01-01

Abstract

Automatic fault type classification is an important ingredient of smart electrical grids. Similar to other machine-learning models, methods developed for fault classification suffer from the issue of lack of transparency. This work sheds light on preliminary insights of an ongoing study, in which we show how feature importance measurement and feature interaction visualization using partial dependence plots (PDPs) can help interpretability of the classification outcomes. While the former, measures the role of each feature on the final predictions in isolation, the latter focuses on mutual interaction between pairs of features. We show the merits of these two complementary feature analysis mechanisms in facilitating interpretability of the fault type classification task.
2020
XAI.it 2020 : Italian Workshop on Explainable Artificial Intelligence 2020 : Proceedings of the Italian Workshop on Explainable Artificial Intelligence co-located with 19th International Conference of the Italian Association for Artificial Intelligence (AIxIA 2020)
CEUR-WS
File in questo prodotto:
File Dimensione Formato  
short8.pdf

accesso aperto

Tipologia: Versione editoriale
Licenza: Creative commons
Dimensione 510.94 kB
Formato Adobe PDF
510.94 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11589/224445
Citazioni
  • Scopus 4
  • ???jsp.display-item.citation.isi??? ND
social impact