Activation of dynamic filter generation for message management systems through gesture-based input
Generation of filtering rules for incoming messages can be activated by a user through a gesture-based input. When the user activates filtering rule generation, the message management service can cause a user device to display a prompt requesting additional data to be used to generate the filtering rule. The gesture can be a swipe and hold gesture, where the swipe is associated with an action (archiving, deleting, deferring, or adding the message to a list), and the swipe and hold action can activate filtering rule generation associated with the action.
1 . A computer-implemented method comprising:
identifying, by a content management system, a plurality of event records associated with a user account, wherein each event record comprises features of content items and user actions taken on the content items;
analyzing, by the content management system, the plurality of event records to detect correlations between the features of the content items and the user actions taken on the content items;
generating, by the content management system, correlation metrics based on the detected correlations between the features of the content items and the user actions taken on the content items, the correlation metrics indicating a degree of correlation between a feature set and a particular user action;
generating, by the content management system, metadata corresponding to a particular correlation metric from the correlation metrics, the metadata defining a rule for processing new content items based on one or more features and one or more actions associated with the particular correlation metric; and
associating, by the content management system, the metadata with a new content item to perform the one or more actions based on determining the new content item comprises the one or more features associated with the particular correlation metric.
2 . The computer-implemented method of claim 1 , further comprising:
identifying a user action, of the user actions, taken on a content item of the content items;
identifying a feature of the features of the content items corresponding to the content item; and
generating an event record based on the first-user action and the feature.
3 . The computer-implemented method of claim 1 , wherein analyzing the plurality of event records further comprises determining a number of consecutive times a user action of the user actions has been taken on a content item of the content items comprising a particular feature of the features of the content items.
4 . The computer-implemented method of claim 1 , wherein analyzing the plurality of event records further comprises detecting, using a machine learning model, a correlation between a subset of the features of the content items and a user action of the user actions taken on the content items.
5 . The computer-implemented method of claim 1 , wherein associating the metadata with the new content item further comprises:
providing, by the content management system to a client device for display, the new content item and a prompt to establish the rule for processing the new content items.
6 . The computer-implemented method of claim 1 , wherein generating the metadata further comprises generating, by the content management system, suggestion metadata defining a suggested rule to automatically perform the one or more actions.
7 . The computer-implemented method of claim 6 , further comprising:
adding, by the content management system, the suggestion metadata to the new content item.
8 . A non-transitory computer readable medium comprising instructions, which, when executed by one or more processors, cause a content management system to:
identify, by the content management system, a plurality of event records corresponding to content items associated with a user account, wherein each event record comprises features of the content items and actions taken on the content items by the user account;
analyze, by the content management system, the plurality of event records to detect correlations between the features of the content items and the actions taken by the user account on the content items;
generate, by the content management system, correlation metrics based on the detected correlations between the features of the content items and the actions taken by the user account on the content items, the correlation metrics indicating a degree of correlation between a feature set and a particular action of the actions taken on the content items;
generate, establishing, by the content management system, metadata corresponding to a particular correlation metric from the correlation metrics, the metadata defining a rule for processing new content items based on one or more features and one or more actions associated with the particular correlation metric; and
associate, by the content management system, the metadata with a new content item to perform the one or more actions based on determining the new content item comprises the one or more features associated with the particular correlation metric.
9 . The non-transitory computer readable medium of claim 8 , further comprising instructions, which, when executed by the one or more processors, cause the content management system to:
identify an action, of the actions taken by the user account, on a content item of the content items;
identifying a feature of the features of the content items corresponding to the content item; and
generating an event record based on the action and the feature.
10 . The non-transitory computer readable medium of claim 8 , further comprising instructions, which, when executed by the one or more processors, cause the content management system to analyze the plurality of event records by determining a number of consecutive times an action of the actions taken by the user account has been taken on a content item comprising a particular feature of the features of content items.
11 . The non-transitory computer readable medium of claim 8 , further comprising instructions, which, when executed by the one or more processors, cause the content management system to analyze the plurality of event records by detecting, using a machine learning model, a correlation between a subset of the features of the content items and an action of the actions taken on the content items by the user account.
12 . The non-transitory computer readable medium of claim 8 , further comprising instructions, which, when executed by the one or more processors, cause the content management system to identify the plurality of event records by identifying a second plurality of event records corresponding to the content items associated with a second user account.
13 . The non-transitory computer readable medium of claim 8 , further comprising instructions, which, when executed by the one or more processors, cause the content management system to:
provide, by the content management system to a client device for display, the new content item and a prompt to establish the rule for processing the new content items.
14 . The non-transitory computer readable medium of claim 13 , further comprising instructions, which, when executed by the one or more processors, cause the content management system to generate the metadata by:
generating, by the content management system, suggestion metadata defining a suggested rule to automatically perform the one or more actions; and
adding, by the content management system, the suggestion metadata to the new content item.
15 . A system, comprising:
one or more processors; and
a memory storing instructions, which, when executed by the one or more processors, causes the system to:
identify a plurality of event records associated with a plurality of user accounts, wherein each event record comprises features of content items and user actions taken on the content items;
analyze the plurality of event records to detect correlations between the features of the content items and the user actions taken on the content items;
generate, based on detecting the correlations, correlation metrics indicating a degree of correlation between a feature set and a particular user action;
generate, metadata corresponding to a particular correlation metric from the correlation metrics, the metadata defining a rule for processing new content items based on one or more features and one or more actions associated with the particular correlation metric; and
associate the metadata with a new content item to perform the one or more actions based on determining the new content item comprises the one or more features associated with the particular correlation metric.
16 . The system of claim 15 , further comprising instructions that, when executed by the one or more processors, cause the system to:
identify a user action, of the user actions, taken on a content item of the content items;
identify a feature of the features of the content items corresponding to the content item; and
generate an event record based on the user action and the feature.
17 . The system of claim 15 , further comprising instructions that, when executed by the one or more processors, cause the system to analyze the plurality of event records by determining a number of consecutive times a user action of the user actions has been taken on a content item of the content items comprising a particular feature of the features of the content items.
18 . The system of claim 15 , further comprising instructions that, when executed by the one or more processors, cause the system to analyze the plurality of event records by detecting, using a machine learning model, a correlation between a subset of the features of the content items and a first-user action of the user actions taken on the content items.
19 . The system of claim 15 , further comprising instructions that, when executed by the one or more processors, cause the system to associate the metadata with the new content items by:
providing, to a client device for display, the new content item and a prompt to establish the rule for processing the new content items.
20 . The system of claim 19 , further comprising instructions that, when executed by the one or more processors, cause the system to generate the metadata by:
generating suggestion metadata defining a suggested rule to automatically perform the one or more actions for; and
adding the suggestion metadata to the new content item.