Taaable 4 is a case-based cooking system which is a contestant in the fourth "Computer Cooking Contest", inheriting most of the features of its previous versions as well as adding new ones. Two new features both concerning knowledge acquisition using formal concept analysis (FCA) are proposed this year. The first feature uses FCA in order to enrich the domain ontology (especially the ingredient hierarchy), making the case retrieval more progressive and more precise. The second feature addresses explicitly the adaptation challenge: given a recipe R and some constraints, how can R be adapted. To compute the best way to adapt R, we propose a FCA-based method for extracting adaptation knowledge. These two knowledge extraction processes exploit additional data (73795 recipes) from the "Recipe Source" database.