Duarte Folgado, presented is work on "4th World Conference on eXplainable Artificial Intelligence".

Duarte Folgado, Senior Scientist at Fraunhofer and CHRC, presented the work "Semantic Concept-Augmented Learning and Post-Hoc Analysis for Time Series Classification" at the 4th World Conference on eXplainable Artificial Intelligence, held in Fortaleza, Brazil (July 1-3).

The work, developed with Matilde Silva, André V. Carreiro (Fraunhofer Portugal AICOS) and Leander Weber (Fraunhofer HHI), shows how introducing human-understandable semantic concepts into deep learning models can make post-hoc explainability methods more intuitive by using the language humans naturally use to describe patterns.

The result is more transparent, trustworthy models, with demonstrated benefits for decision support systems in electrocardiogram interpretation.

hashtag#ExplainableAI hashtag#XAI hashtag#DeepLearning hashtag#Electrocardiogram