US 11,887,014 B2
Dynamic question recommendation
Jeremiah Reeves, Ripon, CA (US); Nithya Rajagopalan, Bangalore (IN); Abhishek Chaturvedi, Bangalore (IN); Sunil Gornalle, Bangalore (IN); Prasad Karani, Karnataka (IN); Surendranath Gopinathan, Bangalore (IN); and Gurudayal Khosla, Bangalore (IN)
Assigned to SAP SE, Walldorf (DE)
Filed by SAP SE, Walldorf (DE)
Filed on Jul. 27, 2018, as Appl. No. 16/047,223.
Prior Publication US 2020/0034720 A1, Jan. 30, 2020
Int. Cl. G06F 16/00 (2019.01); G06N 5/04 (2023.01); G06Q 30/0601 (2023.01); G06N 3/006 (2023.01); G06N 20/00 (2019.01); G06F 16/27 (2019.01); G06F 16/81 (2019.01); G06F 16/838 (2019.01); G06F 16/835 (2019.01)
CPC G06N 5/04 (2013.01) [G06F 16/27 (2019.01); G06F 16/81 (2019.01); G06F 16/838 (2019.01); G06F 16/8373 (2019.01); G06N 3/006 (2013.01); G06N 20/00 (2019.01); G06Q 30/0625 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A system comprising:
a memory; and
a computer-readable medium having instructions stored thereon, which, when executed by a processor, cause the system to perform operations comprising:
receiving a request, via a graphical user interface, to add a new object to a directory of objects, the new object having a first category in a hierarchical taxonomy of categories and objects;
retrieving one or more questions previously assigned prior to the receiving the request, to the first category and/or one or more existing objects within the first category;
feeding each of the retrieved one or more questions and information about the new object into a first machine learned model trained to output a probability that a question is applicable to an object, wherein the first machine learned model is trained by:
feeding one or more sample questions and associated objects into a feature extractor designed to extract one or more features from the one or more sample questions and associated objects; and
passing the extracted one or more features and associated labels into a first machine learning algorithm to train the first machine learned model;
generating one or more questions for the new object based on the probability for each of the retrieved one or more questions;
assigning at least one of the one or more generated questions to the new object; and
causing presentation of one or more of the one or more generated questions to a user for selection.