Intelligent selector control for user interfaces
Methods and systems for intelligently recommending selections for a selector control are disclosed. The method includes receiving a recommendation request from a selector control client, the recommendation request comprising a search string and a unique identifier of a user interacting with a selector control; identifying user identifiers of usernames matching the search string; retrieving machine learning features corresponding to the user identifiers of usernames matching the search string; applying a machine learning model to the retrieved machine learning features to assign weights to the retrieved machine learning features; computing recommendation scores for the user identifiers based on the assigned weights to the retrieved machine learning features; ranking the user identifiers based on the recommendation scores; and forwarding a ranked list of user identifiers to the selector control client for displaying in the selector control for selection by the user interacting with the selector control.
1. A computer-implemented method comprising:
receiving a recommendation request from a selector control client, the recommendation request comprising a search string and a unique identifier of a user interacting with a selector control;
identifying a set of entries matching the search string;
selecting, from a set of candidate machine learning models, a machine learning model for ranking the set of entries, the selecting based at least in part on a type of the selector control;
retrieving, based at least in part on the type of the selector control, machine learning features corresponding to the set of entries matching the search string;
applying the selected machine learning model to the retrieved machine learning features to assign weights to the retrieved machine learning features;
computing recommendation scores for the set of entries based on the assigned weights to the retrieved machine learning features;
ranking the set of entries based on the recommendation scores; and
forwarding a ranked list of the entries to the selector control client for displaying in the selector control for selection by the user interacting with the selector control.
2. The computer-implemented method of claim 1 , wherein:
the selector client control is displayed in a graphical user interface provided by a product platform;
the set of entries correspond to a list of usernames, each username associated with a user account registered with the product platform; and
user selection of a particular candidate entry of the ranked list of entries causes a particular username to be provided as input to the selector control.
3. The computer-implemented method of claim 2 , wherein:
the product platform is an issue tracking platform;
the selector client control is displayed in a graphical user interface provided by the issue tracking platform;
the graphical user interface is an issue view graphical user interface; and
the selector control is an issue assignment control.
4. The computer-implemented method of claim 1 wherein the recommendation request further comprises at least one of a container identifier of a container with which the selector control client is associated or an object identifier of an object with which the selector control client is associated.
5. The computer-implemented method of claim 4 , wherein the machine learning features are retrieved based on at least one of the container identifier or the object identifier.
6. The computer-implemented method of claim 1 , wherein:
the method further comprises periodically updating the machine learning features; and
periodically updating the machine learning features comprises updating the machine learning features based on event records.
7. The computer-implemented method of claim 1 , wherein the machine learning model is at least one of an assignee model, a mention model and a generic model.
8. A computer-implemented method comprising:
in response to a user selection of a selector control on a graphical user interface displayed on a client device, cause generation of a recommendation request, the recommendation request comprising a user identifier and an indicia of a type of selector control that is selected;
based on data received in the recommendation request, including the user identifier, obtain a set of entries using a machine learning model associated with the type of selector designated by the indicia, the machine learning model adapted using a training set of user identifiers and a training set of event records;
determining a ranking of the set of entries based on a set of weights, each weight corresponding to a respective candidate entry of the set of entries;
transmitting the set of entries to the client device;
cause a list of the set of entries to be displayed with respect to the selector control, the list of the set of entries arranged in accordance with the determined ranking; and
in response to a user selection of a particular candidate entry of the list of the set of entries, cause the particular candidate entry to be entered as input to the selector control.
9. The computer-implemented method of claim 8 , wherein:
the graphical user interface is provided by a product platform;
the selector control is a drop-down list control; and
the type of selector control is one of a list including: an assignee control, a mention control, and a generic control.
10. The computer-implemented method of claim 9 , wherein:
the product platform is an issue tracking platform;
the graphical user interface is an issue view graphical user interface; and
the selector control is an issue assignment control.
11. The computer-implemented method of claim 8 , wherein:
the graphical user interface is provided by a product platform;
the training set of user identifiers include user identifiers associated with user accounts registered with the product platform; and
the training set of event records include event records generated in response to user interactions with the product platform.
12. The computer-implemented method of claim 8 , wherein:
the training set of event records are filtered by using a predetermined time period; and
event records occurring outside of the predetermined time period are excluded from the training set of event records.
13. The computer-implemented method of claim 8 , wherein the machine learning model is at least one of an assignee model, a mention model and a generic model.
14. A computer processing system comprising:
one or more processing units;
one or more communication interfaces; and
a non-transitory computer-readable storage medium storing sequences of instructions, which when executed by the one or more processing units, cause the one or more processing units to:
receive a recommendation request from a selector control client, the recommendation request comprising a search string and a unique identifier of a user interacting with a selector control;
identify a set of entries matching the search string;
select, from a set of candidate machine learning models, a machine learning model for ranking the user identifiers, the selecting based at least in part on a type of the selector control;
retrieve, based at least in part on the type of the selector control, machine learning features corresponding to the set of entries matching the search string;
apply the selected machine learning model to the retrieved machine learning features to assign weights to the retrieved machine learning features;
compute recommendation scores for the set of entries based on the assigned weights to the retrieved machine learning features;
rank the set of entries based on the recommendation scores; and
forward a ranked list of the set of entries to the selector control client for displaying in the selector control for selection by the user interacting with the selector control.
15. The computer processing system of claim 14 , wherein:
the selector client control is displayed in a graphical user interface provided by a product platform;
the set of entries correspond to a list of usernames, each username associated with a user account registered with the product platform; and
user selection of a particular candidate entry of the ranked list of entries causes a particular username to be provided as input to the selector control.
16. The computer processing system of claim 15 , wherein:
the product platform is an issue tracking platform;
the graphical user interface is an issue view graphical user interface; and
the selector control is an issue assignment control.
17. The computer processing system of claim 14 , wherein the recommendation request further comprises at least one of a container identifier of a container with which the selector control client is associated or an object identifier of an object with which the selector control client is associated.
18. The computer processing system of claim 17 , wherein the machine learning features are retrieved based on at least one of the container identifier or the object identifier.
19. The computer processing system of claim 14 , wherein the machine learning model is at least one of an assignee model, a mention model and a generic model.
20. The computer processing system of claim 14 , wherein user selection of a particular candidate entry of the ranked list of the set of entries is used as feedback for the machine learning model.