Explaining Machine Learning Models of Emotion Using the BIRAFFE Dataset

Bobek, Szymon; Tragarz, Magdalena M.; Szelążek, Maciej; Nalepa, Grzegorz J.

doi:10.1007/978-3-030-61534-5_26

Cited by 1 publication

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both [28,29] use Shapley values to explain their models on sentiment analysis, although this analysis is in a different context than FER. Prajod et al used LRP saliency maps to investigate whether a network has learned concepts (in this case action units), especially in the case where a network originally trained for emotion recognition is used as a base for transfer-learning a model to recognize pain [30].…”

Section: Xai In Affective Computing and Fermentioning

confidence: 99%

Comparing Approaches for Explaining DNN-Based Facial Expression Classifications

Burg

Kaya

2022

Algorithms

View full text Add to dashboard Cite

Classifying facial expressions is a vital part of developing systems capable of aptly interacting with users. In this field, the use of deep-learning models has become the standard. However, the inner workings of these models are unintelligible, which is an important issue when deploying them to high-stakes environments. Recent efforts to generate explanations for emotion classification systems have been focused on this type of models. In this work, an alternative way of explaining the decisions of a more conventional model based on geometric features is presented. We develop a geometric-features-based deep neural network (DNN) and a convolutional neural network (CNN). Ensuring a sufficient level of predictive accuracy, we analyze explainability using both objective quantitative criteria and a user study. Results indicate that the fidelity and accuracy scores of the explanations approximate the DNN well. From the performed user study, it becomes clear that the explanations increase the understanding of the DNN and that they are preferred over the explanations for the CNN, which are more commonly used. All scripts used in the study are publicly available.

show abstract

Section: Xai In Affective Computing and Fermentioning

confidence: 99%