US 11,748,419 B2
Generating and executing an optimal dialogue strategy
Yunfeng Zhang, Tarrytown, NY (US); Vera Liao, White Plains, NY (US); and Biplav Srivastava, Rye, NY (US)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Aug. 17, 2021, as Appl. No. 17/404,150.
Application 17/404,150 is a continuation of application No. 16/192,204, filed on Nov. 15, 2018, granted, now 11,138,272.
Prior Publication US 2021/0374181 A1, Dec. 2, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/00 (2019.01); G06F 16/907 (2019.01); G06N 20/00 (2019.01); G06F 16/332 (2019.01); G06F 16/9032 (2019.01); H04L 67/10 (2022.01)
CPC G06F 16/907 (2019.01) [G06F 16/3329 (2019.01); G06F 16/90332 (2019.01); G06N 20/00 (2019.01); H04L 67/10 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory that stores computer executable components;
a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
a dialog strategy component that iteratively performs, while a user is engaged in a dialog with an information system:
trains a machine learning model to learn items and features associated with user dialogues for generating and selecting questions to present to the user, the training comprising:
obtains, by the machine learning model, dialogue input data from the user;
generates, by the machine learning model, an open-ended question and a closed-ended question based on the dialogue input data; and
selects, by the machine learning model, dialogue one of the open-ended question or the closed-ended question that produces a highest amount of change of entropy associated with information gain for presentation by the information system in response to the dialogue input data.