US 11,989,515 B2
Adjusting explainable rules using an exploration framework
Prithviraj Sen, San Jose, CA (US); Yiwei Yang, Flushing, NY (US); Yunyao Li, San Jose, CA (US); and Eser Kandogan, Mountain View, CA (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Feb. 28, 2020, as Appl. No. 16/805,470.
Prior Publication US 2021/0271817 A1, Sep. 2, 2021
Int. Cl. G06F 40/289 (2020.01); G06F 40/35 (2020.01); G06N 3/08 (2023.01)
CPC G06F 40/289 (2020.01) [G06F 40/35 (2020.01); G06N 3/08 (2013.01)] 22 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
creating a plurality of linguistic expressions utilizing a deep neural network (DNN) that has been trained with training data, each of the linguistic expressions having one or more conditions therein;
changing one or more of the conditions of the plurality of linguistic expressions to create an updated plurality of linguistic expressions;
translating each of the updated linguistic expressions into a predetermined query language; and
constructing a sentence classification model, utilizing the translated linguistic expressions,
wherein changing the one or more conditions of the plurality of linguistic expressions includes altering, via a visual exploration framework, a wording of at least one condition of the plurality of linguistic expressions to create an updated condition.