Personalization of content suggestions for document creation
The present disclosure relates to processing operations that generate and present personalized content suggestions to assist a user with document creation. Machine learning modeling may be trained and implemented to evolve pre-canned suggestions for document creation into highly personalized content suggestions, thereby improving the document creation process and user interface experience for users of applications/services that are utilized to create digital documents. As an example, signal data may be detected and analyzed, identifying a specific user's intent to create a digital document. Machine learning modeling may be implemented to evaluate different aspects of collected signal data and identify content from previously created documents, associated with a user account, that may be most relevant to the real-time document creation experience of the user. Personalized contextual suggestions may be presented to a user through a user interface. Examples described herein may be extensible across any type of application/service configured for document creation.
1. A method comprising:
receiving, through a user interface of a productivity application or service, a request for template creation that comprises an identification of a template type for creation of a template;
evaluating signal data associated with the request, wherein an evaluation of the signal data comprises identifying the template type and a specific user account associated with the request;
generating a personalized content suggestion for inclusion in the template by applying a trained machine learning model that is configured to generate the personalized content suggestion based on an evaluation of content from previously created documents that were created by the specific user account and have a document type matching the template, wherein the trained machine learning model executes processing operations that comprise:
filtering, from stored documents, a subset of one or more previously created documents that were created by the specific user account and have the document type matching the template, and
generating the personalized content suggestion based on analysis of content of the subset of one or more previously created documents; and
presenting, through the user interface, the personalized content suggestion in a surfaced representation of the template.
2. The method of claim 1 , further comprising: receiving, through the user interface, a selection of the personalized content suggestion; and generating an electronic document that comprises one or more content portions associated with the personalized content selection selected.
3. The method of claim 1 , wherein the presenting presents the personalized content suggestion in addition to pre-canned design suggestions usable for template generation.
4. The method of claim 1 , wherein the receiving of the request for document creation comprises receiving a selection of the template type; and receiving textual input from the user account for template creation, wherein the evaluation of the signal data comprises identifying the textual input, and wherein the generating of the personalized content suggestion comprises matching the textual input with content from the previously created documents.
5. The method of claim 1 , wherein the filtering of the subset of one or more previously created documents occurs based on a result of a relevance ranking, generated by the trained machine learning model, that correlates respective document types of the plurality of documents with the template type.
6. The method of claim 5 , wherein the trained machine learning model, in generating the personalized content suggestion, executes processing operations that comprise: generating a user-specific relevance ranking of content portions of the subset of one or more previously created documents relative to the template type based on analysis of user-specific signal data that comprises signal data indicating usage patterns of a specific content portion appearing across the subset of one or more of the previously created documents, and wherein the presenting of the personalized content suggestion comprises selecting, from a plurality of personalized content suggestions, the personalized content suggestion based on a result of analyzing the user-specific relevance ranking.
7. The method of claim 5 , wherein the trained machine learning model, in generating the personalized content suggestion, executes processing operations that comprise: generating a user-specific relevance ranking of content portions of the subset of one or more previously created documents relative to the template type based on analysis of user-specific signal data that comprises signal data indicating user actions associated with specific content portions of the previously created documents, and wherein the presenting of the personalized content suggestion comprises selecting, from a plurality of personalized content suggestions, the personalized content suggestion based on a result of analyzing the user-specific relevance ranking.
8. The method of claim 1 , wherein filtering of the subset of one or more previously created documents that were created by the specific user account further occurs based on a user access relevance analysis that correlates user access by the specific user account to a previously created document, and wherein a threshold time period is set for the user access.
9. A system comprising:
at least one processor; and
a memory, operatively connected with the at least one processor, storing computer- executable instructions that, when executed by the at least one processor, causes the at least one processor to execute a method that comprises:
receiving, through a user interface of a productivity application or service, a request for template creation that comprises an identification of a template type for creation of a template;
evaluating signal data associated with the request, wherein an evaluation of the signal data comprises identifying the template type and a specific user account associated with the request;
generating a personalized content suggestion for inclusion in the template by applying a trained machine learning model that is configured to generate the personalized content suggestion based on an evaluation of content from previously created documents that were created by the specific user account and have a document type matching the template, wherein the trained machine learning model executes processing operations that comprise:
filtering, from stored documents, a subset of one or more previously created documents that were created by the specific user account and have the document type matching the template, and
generating the personalized content suggestion based on analysis of content of the subset of one or more previously created documents; and
presenting, through the user interface, the personalized content suggestion in a surfaced representation of the template.
10. The system of claim 9 , wherein the method, executed by the at least one processor, further comprises: receiving, through the user interface, a selection of the personalized content suggestion; and generating an electronic document that comprises one or more content portions associated with the personalized content selection selected.
11. The system of claim 9 , wherein the receiving of the request for document creation comprises receiving a selection of the template type; and receiving textual input from the user account for template creation, wherein the evaluation of the signal data comprises identifying the textual input, and wherein the generating of the personalized content suggestion comprises matching the textual input with content from the previously created documents.
12. The system of claim 9 , wherein the filtering of the subset of one or more previously created documents occurs based on a result of a relevance ranking, generated by the trained machine learning model, that correlates respective document types of the plurality of documents with the template type.
13. The system of claim 12 , wherein the trained machine learning model, in generating the personalized content suggestion, executes processing operations that comprise: generating a user-specific relevance ranking of content portions of the subset of one or more previously created documents relative to the template type based on analysis of user-specific signal data that comprises signal data indicating user actions associated with specific content portions of the previously created documents, and wherein the presenting of the personalized content suggestion comprises selecting, from a plurality of of the personalized content suggestions, the personalized content suggestion based on a result of analyzing the user-specific relevance ranking.
14. The system of claim 12 , wherein the filtering of the subset of one or more previously created documents occurs based on a result of a relevance ranking, generated by the trained machine learning model, that correlates respective document types of the plurality of documents with the template type.
15. The system of claim 9 , wherein filtering of the subset of one or more previously created documents that were created by the specific user account further occurs based on a user access relevance analysis that correlates user access by the specific user account to a previously created document, and wherein a threshold time period is set for the user access.
16. A computer-readable storage media storing computer-executable instructions that, when executed by at least one processor, causes the at least one processor to execute a method comprising:
receiving, through a user interface of a productivity application or service, a request for template creation that comprises an identification of a template type for creation of a template;
evaluating signal data associated with the request, wherein an evaluation of the signal data comprises identifying the template type and a specific user account associated with the request;
generating a personalized content suggestion for inclusion in the template by applying a trained machine learning model that is configured to generate the personalized content suggestion based on an evaluation of content from previously created documents that were created by the specific user account and have a document type matching the template, wherein the trained machine learning model executes processing operations that comprise:
filtering, from stored documents, a subset of one or more previously created documents that were created by the specific user account and have the document type matching the template, and
generating the personalized content suggestion based on analysis of content of the subset of one or more previously created documents; and
transmitting, to a client device, data for rendering the personalized content suggestion in a surfaced representation of the template.
17. The computer-readable storage media of claim 16 , wherein the filtering of the subset of one or more previously created documents occurs based on a result of a relevance ranking, generated by the trained machine learning model, that correlates respective document types of the plurality of documents with the template type.
18. The computer-readable storage media of claim 17 , wherein the trained machine learning model, in generating the personalized content suggestion, execute executes processing operations that comprise: generating a user-specific relevance ranking of content portions of the subset of one or more previously created documents relative to the template type based on analysis of user-specific signal data that comprises signal data indicating usage patterns of a specific content portion appearing across the subset of one or more of the previously created documents, and wherein the presenting of the personalized content suggestion comprises selecting, from a plurality of personalized content suggestions, the personalized content suggestion based on a result of analyzing the user-specific relevance ranking.
19. The computer-readable storage media of claim 17 , wherein the trained machine learning model, in generating the personalized content suggestion, executes processing operations that comprise: generating a user-specific relevance ranking of content portions of the subset of one or more previously created documents relative to the template type based on analysis of user-specific signal data that comprises signal data indicating user actions associated with specific content portions of the previously created documents, and wherein the presenting of the personalized content suggestion comprises selecting, from a plurality of personalized content suggestions, the personalized content suggestion based on a result of analyzing the user-specific relevance ranking.
20. The computer-readable storage media of claim 16 , wherein filtering of the subset of one or more previously created documents that were created by the specific user account further occurs based on a user access relevance analysis that correlates user access by the specific user account to a previously created document, and wherein a threshold time period is set for the user access.