Systems and methods for providing feed page updates in a social networking system

A system comprising:

• at least one hardware processor;
• and
• a memory storing instructions that, when executed by the at least one processor, cause the system to perform:
• determining one or more user-related metrics related to of a plurality of pages associated with an administrator based at least in part on a likelihood of the administrator selecting the one or more user-related metrics, wherein the likelihood of the administrator selecting the one or more user-related metrics is determined by a first machine learning model and the first machine learning model is trained based at least in part on tracked information about user-related metrics and pages with which page administrators have engaged;
• determining one or more recommendations related to the plurality of pages based at least in part on a likelihood of the administrator selecting the one or more recommendations and a number of determined user-related metrics, wherein the likelihood of the administrator selecting the one or more recommendations is determined by a second machine learning model and the second machine learning model is trained based at least in part on the tracked information about user-related metrics and page recommendations selected by the page administrators;
• and
• determining one or more pages of the plurality of pages for which to provide one or more cards of page updates for the administrator, based at least in part on the determined user-related metrics and the determined recommendations;
• and
• providing a content item that includes the one or more cards, wherein the one or more cards include the determined user-related metrics and the determined recommendations for the one or more pages corresponding to the one or more cards and wherein a selection of at least one of:
• the one or more cards, the determined user-related metrics, or the determined recommendations is included in new training data for further training at least one of:
• the first machine learning model or the second machine learning model.

A non-transitory computer readable medium including instructions that, when executed by at least one hardware processor of a computing system, cause the computing system to perform a method comprising:

• determining one or more user-related metrics related to a plurality of pages associated with an administrator based at least in part on a likelihood of the administrator selecting the one or more user-related metrics, wherein the likelihood of the administrator selecting the one or more user-related metrics is determined by a first machine learning model and the first machine learning model is trained based at least in part on tracked information about user-related metrics and pages with which page administrators have engaged;
• determining one or more recommendations related to the plurality of pages based at least in part on a likelihood of the administrator selecting the one or more recommendations and a number of determined user-related metrics, wherein the likelihood of the administrator selecting the one or more recommendations is determined by a second machine learning model and the second machine learning model is trained based at least in part on the tracked information about user-related metrics and page recommendations selected by the page administrators;
• and
• determining one or more pages of the plurality of pages for which to provide one or more cards of page updates for the administrator, based at least in part on the determined user-related metrics and the determined recommendations;
• and
• providing a content item that includes the one or more cards, wherein the one or more cards include the determined user-related metrics and the determined recommendations for the one or more pages corresponding to the one or more cards and wherein a selection of at least one of:
• the one or more cards, the determined user-related metrics, or the determined recommendations is included in new training data for further training at least one of:
• the first machine learning model or the second machine learning model.