L. Schwenke and M. Atzmueller, “Making Time Series Embeddings More Interpretable in Deep Learning: Extracting Higher-Level Features via Symbolic Approximation Representations”, FLAIRS, vol. 36, no. 1, May 2023.