Fig. 2

Schematic representation of the Logic Learning Machine algorithm. In the first phase (Latticization) each variable is transformed into a string of binary data, using the inverse only-one code binarization and all strings are eventually concatenated in one unique large string per each subject. In the second phase (Shadow Clustering) a set of binary vectors (the “implicants”) is generated, each of which identifies a cluster in the input space associated with a specific output class. Finally, all the implicants are transformed into simple conditions and combined in a set of intelligible rules