公开(公告)号：US10884718B2

公开(公告)日：2021-01-05

申请号：US15779916

申请日：2016-11-30

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Anand Srinivasan Natesan ,  Rithesh Sreenivasan ,  Rajendra Singh Sisodia ,  Karthik Srinivasan ,  Debasish Chatterjee

IPC: G06F8/38 ,  G06N20/00 ,  G06Q10/06 ,  G16H10/20 ,  G16H20/00

Abstract: The present invention relates to a device (10) for use in improving a user interaction with a user interface application. It is described to provide (210) data for a user, the data comprising interaction data of the user with a plurality of interaction fields of a first version of a user interface application. The interaction data is clustered (220) into a plurality of groups. The interaction data is analysed (230) to determine the contribution a specific user interaction field of the plurality of interaction fields makes to a specific group of the plurality of groups. The specific user interaction field is modified (240) on the basis of the determined contribution the specific user interaction field makes. A second version of the user interface application is displayed (250) comprising the modified specific user interaction field.

公开(公告)号：US11074485B2

公开(公告)日：2021-07-27

申请号：US16451237

申请日：2019-06-25

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Rithesh Sreenivasan ,  Aart Tijmen Van Halteren ,  Karthik Srinivasan

IPC: G06K9/62 ,  G06N5/04 ,  G06N20/00

Abstract: A machine learning based recommendation model, including a supervised learning classifier configured to receive input training data that includes a plurality of behavioral determinants, a supervised learning model configured to receive subject input data that includes a plurality of behavior determinants, wherein the supervised learning model outputs a predicted behavior of a subject, and a channel selection module configured to receive the subject input data and the predicted behavior and to determine a recommended communication channel for the subject to follow to achieve the predicted behavior.