Figure 1From: Extracting complements and substitutes from sales data: a network perspectiveSchematic diagram showing the data structure of the sales data in Sect. 2.1 and the flavour compound and recipe data in Sect. 2.3, and their corresponding bipartite networks, together with the matching between the products and the ingredients, where the “Matching” step is required because of the different names that can appear in different datasetsBack to article page