Higher-order networks are a class of graphs that incorporate “memory no-des” in order to take into account the indirect interactions that can existin sequential data. They differ from so-called “order 1” networks, whichonly take direct relationships into account. In this article, we provide anoverview of this concept, detailing their construction and the mining tech-niques that can be employed. We present the Python package honyx, whichcontains algorithms already available in the literature. We propose a tuto-rial on its use through a case study of commercial flight itineraries in theUnited States. We also discuss some of the challenges and future directionsin the field.
