Exploring the Relationship Between Feature Attribution Methods and Model Performance

Autor(es)

Silva, Priscylla ; Silva, Claudio ; Nonato, Luis Gustavo

• Tipo de documento: Artigo Científico
• Fonte: NEURIPS WORKSHOPS, 2020; v. 257, p. 10-pg., 2024-01-01.

Resumo

Machine learning and deep learning models are pivotal in educational contexts, particularly in predicting student success. Despite their widespread application, a significant gap persists in comprehending the factors influencing these models' predictions, especially in explainability within education. This work addresses this gap by employing nine distinct explanation methods and conducting a comprehensive analysis to explore the correlation between the agreement among these methods in generating explanations and the predictive model's performance. Applying Spearman's correlation, our findings reveal a very strong correlation between the model's performance and the agreement level observed among the explanation methods. (AU)

• Processo FAPESP: 23/05783-5 - Investigando o problema de desacordo em métodos de explicação locais
• Beneficiário: Priscylla Maria da Silva Sousa
• Modalidade de apoio: Bolsas no Exterior - Estágio de Pesquisa - Doutorado
• Processo FAPESP: 22/09091-8 - Criminalidade, insegurança e legitimidade: uma abordagem transdisciplinar
• Beneficiário: Luis Gustavo Nonato
• Modalidade de apoio: Auxílio à Pesquisa - Programa eScience e Data Science - Temático
• Processo FAPESP: 22/03941-0 - Um modelo de predição de crimes interpretável usando Redes Neurais para Grafos
• Beneficiário: Priscylla Maria da Silva Sousa
• Modalidade de apoio: Bolsas no Brasil - Doutorado