US 11,989,667 B2
Interpretation of machine leaning results using feature analysis
Yann Le Biannic, Suresnes (FR)
Assigned to BUSINESS OBJECTS SOFTWARE LTD., Dublin (IE)
Filed by Business Objects Software Ltd, Dublin (IE)
Filed on Jun. 5, 2023, as Appl. No. 18/206,020.
Application 18/206,020 is a continuation of application No. 16/712,792, filed on Dec. 12, 2019, granted, now 11,727,284.
Prior Publication US 2023/0316111 A1, Oct. 5, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06N 5/04 (2013.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computing system comprising:
memory;
one or more processing units coupled to the memory; and
one or more computer readable storage media storing instructions that, when executed, cause the computing system to perform operations comprising:
receiving a training data set, the training data set comprising values for a first plurality of features;
training a machine learning algorithm using the training data set to provide a trained machine learning model
processing an analysis data set using the trained machine learning model to provide a result;
forming a plurality of feature groups, at least one of the feature groups comprising a second plurality of features of the first plurality of features, the second plurality of features being a proper subset of the first plurality of features, the forming a plurality of feature groups comprising determining contextual contributions of at least a portion of the first plurality of features to the result and detecting causality cycles within feature groups; and
in response to determining contextual contributions and detecting causality cycles, determining which features and feature groups to employ in retraining the machine learning model.