WHALE (Web-Scale Hybrid Explainable Machine Learning) aims at developing scalable, transparent, and explainable AI systems for web-scale knowledge graphs. Building on advanced machine learning and semantic web technologies, WHALE focuses on enhancing the explainability of AI through novel class expression learning (CEL) techniques that combine traditional and cutting-edge methodologies.

Recognizing the challenges posed by the vast size and complexity of web data, WHALE employs hybrid machine learning approaches that integrate embeddings and tensor representations to process and analyze massive datasets efficiently. The project aims to improve the speed and scalability of AI models, making them more practical for real-world applications while maintaining rigorous standards of transparency and interpretability.