Leveraging Knowledge Graph to design the Machine-Learning Engineering Body-of-Knowledge

Juliette Mattioli (1) , Dominique Tachet (2) , Fabien Tschirhart (2) , Henri Sohier (2) , Loic Cantat (2) , Boris Robert (3)

1 THALES [France]
2 IRT SystemX
3 IRT Saint Exupéry - Institut de Recherche Technologique

A body of knowledge (BoK) is the complete set of concepts, terms, standards and activities promoting abroad awareness of a field or profession to guide practice or work. This paper presents how knowledge-based artificial intelligence (a.k.a. symbolic AI) could be used to build a body of knowledge (BoK) by first identifying relevant documents and data to capture concepts, standards, best practices, and state-of-the-art; then fusing all knowledge items into a knowledge graph, and finally providing query capacities. The overall process of knowledge collection, storage, and retrieval is implemented to support a trustworthy Machine Learning end-to-end engineering

Mots clés: Body of knowledge; Knowledge graph; Knowledge extraction; Knowledge fusion; Machine Learning; Body-of-Knowledge; ML Engineering

Domaines: Intelligence artificielle [cs.AI]; Machine Learning [stat.ML]

Juliette Mattioli, Dominique Tachet, Fabien Tschirhart, Henri Sohier, Loic Cantat, et al.. Leveraging Knowledge Graph to design the Machine-Learning Engineering Body-of-Knowledge. IEEE International Conference on AI x Science, Technology, and Technology (AIxSET), Sep 2024, Laguna hills, United States. ⟨hal-04706669⟩