User interface generation with machine learning

Patent metadata

Field	Value
Publication number	US-11625141-B2
Application number	US-202017028474-A
Country	US
Kind code	B2
Filing date	Sep 22, 2020
Priority date	Sep 22, 2020
Publication date	Apr 11, 2023
Grant date	Apr 11, 2023

Abstract

A system includes a processor configured to perform operations, including receiving, from a client device, a request, associated with a user identifier, for rendering of a graphical user interface (GUI), and obtaining, based on the request and from persistent storage, a framework definition that specifies a first plurality of components and a layout thereof within the GUI. The operations also include determining a second plurality of components by way of a machine learning model and based on attributes associated with the user identifier, and determining, for each respective component of the second plurality of components, a corresponding visual format and a corresponding position within the GUI. The operations further include updating the framework definition by combining the first plurality of components and the second plurality of components based on the corresponding visual format and the corresponding position, and transmitting the framework definition as updated to the client device.

First claim

What is claimed is:

1. A system comprising:
a persistent storage containing a framework definition that specifies a first plurality of predetermined components of a graphical user interface and a layout of the first plurality of predetermined components within the graphical user interface, wherein each respective predetermined component of the first plurality of predetermined components is associated with a corresponding software application and occupies a corresponding frame that forms a bounded portion of the graphical user interface;
a processor configured to perform operations comprising:
receiving, from a client device, a request for rendering of the graphical user interface, wherein the request is associated with a user identifier;
obtaining, based on receiving the request and from the persistent storage, the framework definition;
determining, by way of a machine learning model, a second plurality of components comprising, for each respective predetermined component of the first plurality of predetermined components, corresponding one or more components, wherein the machine learning model is configured to generate the corresponding one or more components based on (i) one or more attributes associated with the user identifier, (ii) an identifier of the respective predetermined component, (iii) data representing one or more states of the corresponding software application, and (iv) at least one attribute of at least one component generated in connection with one or more other user identifiers belonging to a same group as the user identifier, wherein the corresponding one or more components are generated by the machine learning model to summarize the one or more states of the corresponding software application by displaying a corresponding data subset selected by the machine learning model from the data representing the one or more states;
determining, for each respective component of the second plurality of components, (i) a corresponding visual format according to which data selected for the respective component is displayed and (ii) a corresponding position of the respective component within the corresponding frame of a corresponding predetermined component of the first plurality of predetermined components;
updating the framework definition by combining the first plurality of predetermined components and the second plurality of components based on (i) the corresponding visual format determined for each respective component of the second plurality of components and (ii) the corresponding position determined for each respective component of the second plurality of components; and
transmitting the framework definition as updated to the client device.

2. The system of claim 1 , wherein updating the framework definition by combining the first plurality of predetermined components and the second plurality of components comprises:
for each respective predetermined component of the first plurality of predetermined components, incorporating the corresponding one or more components into the respective predetermined component, wherein the respective predetermined component is configured to expand to display the corresponding one or more components in response to selection of the respective predetermined component by way of the client device.

3. The system of claim 1 , wherein: (i) the framework definition further comprises a predetermined favorites component that represents a favorites section of the graphical user interface and the machine learning model is configured to generate, based on the one or more attributes and an identifier of the predetermined favorites component, one or more favorites components determined to be more frequently used in connection with the user identifier than other candidate components of the graphical user interface or (ii) the framework definition further comprises a predetermined recommended component that represents a recommended section of the graphical user interface and the machine learning model is configured to generate, based on the one or more attributes associated with the user identifier and an identifier of the predetermined recommended component, one or more recommended components that represent recommendations made for the user identifier by the machine learning model.

4. The system of claim 1 , wherein the corresponding visual format according to which data selected for the respective component is displayed is determined based on one or more prior visual formats previously used for the respective component in a prior rendering of the graphical user interface in connection with the user identifier, and wherein the corresponding position of the respective component within the corresponding predetermined component is determined based on one or more prior positions previously used for the respective component in the prior rendering of the graphical user interface in connection with the user identifier.

5. The system of claim 1 , wherein the corresponding visual format according to which data selected for the respective component is displayed and the corresponding position of the respective component within the corresponding frame of the corresponding predetermined component are each determined by way of a second machine learning model trained to generate the corresponding visual format and the corresponding position based on one or more of: (i) the one or more attributes associated with the user identifier, (ii) one or more attributes of the corresponding predetermined component of the first plurality of predetermined components, (iii) one or more attributes of the data selected for the respective component, or (iv) the at least one attribute of the at least one component generated in connection with the one or more other user identifiers belonging to the same group as the user identifier.

6. The system of claim 1 , wherein the operations further comprise:
receiving, from the client device, an indication of a manual modification to one or more components of the second plurality of components;
generating, based on the indication of the manual modification, training data for the machine learning model, wherein the training data comprises (i) a representation of the graphical user interface following the manual modification and (ii) the one or more attributes associated with the user identifier, and wherein the machine learning model is configured to be retrained based on the training data; and
storing, in the persistent storage, the training data.

7. The system of claim 1 , wherein the one or more attributes associated with the user identifier comprise one or more of: (i) one or more user preferences associated with the user identifier, (ii) an application usage history associated with the user identifier, (iii) a role associated with the user identifier, (iv) a geographic region associated with the user identifier, or (v) a performance evaluation score associated with the user identifier.

8. The system of claim 1 , wherein the machine learning model is configured to select the corresponding data subset from the data representing the one or more states further based on one or more attributes associated with the one or more states as represented by the data.

9. The system of claim 1 , wherein the corresponding visual format comprises at least one of: (i) a card, (ii) a list, (iii) a graph, (iv) a table, (v) a text field, (vi) a modal window, or (vii) an overlay.

10. The system of claim 1 , wherein the framework definition indicates a predetermined grouping of a subset of the first plurality of predetermined components and includes an overview component representing the predetermined grouping, wherein selection of the overview component by way of the client device is configured to cause

Assignees

• Servicenow Inc

Inventors

• Azmoon Troy
• Wade Aaron
• Castillo Rita Maria

Classifications

• G06F16/9538 Primary

  Presentation of query results · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06F3/0482

  Interaction with lists of selectable items, e.g. menus · CPC title

• G06F9/451

  Execution arrangements for user interfaces · CPC title

• G06N20/00

  Machine learning · CPC title