US 11,734,582 B2
Automated rule generation framework using machine learning for classification problems
Atreya Biswas, Singapore (SG); and Srivatsan Santhanam, Bangalore (IN)
Assigned to SAP SE, Walldorf (DE)
Filed by SAP SE, Walldorf (DE)
Filed on Oct. 31, 2019, as Appl. No. 16/669,686.
Prior Publication US 2021/0133515 A1, May 6, 2021
Int. Cl. G06F 3/048 (2013.01); G06N 5/025 (2023.01); G06Q 10/10 (2023.01); G06N 20/10 (2019.01); G06F 18/243 (2023.01); G06N 5/01 (2023.01)
CPC G06N 5/025 (2013.01) [G06F 18/24323 (2023.01); G06N 5/01 (2023.01); G06N 20/10 (2019.01); G06Q 10/10 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for automated provision of sets of rules using machine learning (ML) models, the method being executed by one or more processors and comprising:
receiving historical data, the historical data comprising a set of variable vectors, each variable vector being assigned to a class in a set of classes;
processing the historical data through one or more encoders to provide a set of feature vectors, each feature vector corresponding to a respective variable vector in the set of variable vectors and being assigned to the class of the respective variable vector;
generating a set of decision trees based on the feature vectors, the set of decision trees comprising a first decision tree corresponding to a first class in the set of classes and a second decision tree corresponding to a second class in the set of classes;
transforming the first decision tree into a first set of rules and the second decision tree into a second set of rules to provide sets of rules comprising the first set of rules and the second set of rules, each rule in the first set of rules defining one or more if conditions to selectively assign at least a portion of an electronic document to the first class, and each rule in the second set of rules defining one or more if conditions to selectively assign the at least a portion of the electronic document to the second class; and
providing the sets of rules for execution in an enterprise system, the enterprise system classifying electronic documents to classes in the set of classes based on the sets of rules.