Comparison of feature importance measures as explanations for classification models
| Autor: | Susanne Jauhiainen, Mirka Saarela |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
feature importance
Computer science General Chemical Engineering General Physics and Astronomy 02 engineering and technology interpretable models tekoäly Machine learning computer.software_genre Logistic regression Domain (software engineering) 020204 information systems 0202 electrical engineering electronic engineering information engineering Feature (machine learning) General Materials Science General Environmental Science luokitus (toiminta) explainable artificial intelligence business.industry logistic regression General Engineering Random forest koneoppiminen Trustworthiness Injury data General Earth and Planetary Sciences 020201 artificial intelligence & image processing Artificial intelligence business computer random forest |
| Zdroj: | SN Applied Sciences. 3 |
| ISSN: | 2523-3971 2523-3963 |
| DOI: | 10.1007/s42452-021-04148-9 |
| Popis: | Explainable artificial intelligence is an emerging research direction helping the user or developer of machine learning models understand why models behave the way they do. The most popular explanation technique is feature importance. However, there are several different approaches how feature importances are being measured, most notably global and local. In this study we compare different feature importance measures using both linear (logistic regression with L1 penalization) and non-linear (random forest) methods and local interpretable model-agnostic explanations on top of them. These methods are applied to two datasets from the medical domain, the openly available breast cancer data from the UCI Archive and a recently collected running injury data. Our results show that the most important features differ depending on the technique. We argue that a combination of several explanation techniques could provide more reliable and trustworthy results. In particular, local explanations should be used in the most critical cases such as false negatives. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink