Method, medium, and system for a user interface with search results logically organized by carousels

A computer program product comprising a non-transitory computer readable storage medium having instructions encoded thereon that, when executed by a processor, cause the processor to:

• store user data describing a user of an online system, wherein the user data describes a plurality of interactions of the user with the online system;
• receive a search query at the online system from a client device associated with the user;
• responsive to receiving the search query, generate a plurality of carousels to display on the client device as a response to the search query, wherein each of the plurality of carousels is horizontally scrollable and comprises content describing a plurality of items, wherein generating the plurality of carousels comprises:
• computing a query relevance probability score for each of a plurality of items stored in an item database based on the search query, the stored user data, and item data describing the plurality of items, wherein the query relevance probability score for an item is a score representing a predicted likelihood that a user would interact with the item when the item is presented to the user in response to a search query from the user;
• identifying a set of matching items from the plurality of items based on the computed query relevance probability scores for the plurality of items;
• obtaining, from a hierarchical item taxonomy, a taxonomy category label for each matching item in the set of matching items, wherein the hierarchical item taxonomy relates each item in the item database to one of a plurality of taxonomy category labels;
• grouping items from the set of matching items based on the taxonomy category labels obtained from the hierarchical item taxonomy, wherein the grouping produces a plurality of groups, and wherein each of the plurality of groups comprises items from the set of matching items that are associated with a branch within the hierarchical item taxonomy associated with the category label associated with the group;
• and
• generating a carousel for each of the plurality of groups, wherein a carousel for a group is a scrollable user interface element comprising items associated with the group arranged in a list;
• automatically arrange the generated carousels into a scrollable list based on the stored user data to generate a user interface displaying the set of matching items,
• wherein the scrollable list is vertically scrollable and wherein automatically arranging the generated carousels comprises:
• applying a machine-learned carousel model to the stored user data, item data describing the set of matching items, and the received search query to generate a score for each of the generated carousels, wherein a generated score for a carousel represents a predicted likelihood that the branch of the hierarchical item taxonomy is relevant to the search query, and
• automatically arranging the generated carousels based on the generated scores, wherein the generated carousels are automatically arranged such that generated carousels with generated scores indicating a higher predicted likelihood are arranged vertically-higher in the scrollable list than generated carousels with generated scores indicating a lower predicted likelihood;
• and
• send instructions for the user interface to the client device for display, wherein sending the instructions for the user interface to the client device causes the client device to automatically display the scrollable list of the plurality of carousels as results to the received search query, wherein the scrollable list of the plurality of carousels are displayed according to the automatic arrangement of the generated carousels, and wherein each of the generated carousels are displayed in the user interface with text describing the taxonomy category label associated with the generated carousel.

A system, comprising:

• a processor;
• and
• a non-transitory computer readable storage medium having instructions encoded thereon that, when executed by the processor, cause the processor to:
• store user data describing a user of an online system, wherein the user data describes a plurality of interactions of the user with the online system;
• receive a search query at the online system from a client device associated with the user;
• responsive to receiving the search query, generate a plurality of carousels to display on the client device as a response to the search query, wherein each of the plurality of carousels is horizontally scrollable and comprises content describing a plurality of items, wherein generating the plurality of carousels comprises:
• computing a query relevance probability score for each of a plurality of items stored in an item database based on the search query, the stored user data, and item data describing the plurality of items, wherein the query relevance probability score for an item is a score representing a predicted likelihood that a user would interact with the item when the item is presented to the user in response to a search query from the user;
• identifying a set of matching items from the plurality of items based on the computed query relevance probability scores for the plurality of items;
• obtaining, from a hierarchical item taxonomy, a taxonomy category label for each matching item in the set of matching items, wherein the hierarchical item taxonomy relates each item in the item database to one of a plurality of taxonomy category labels;
• grouping items from the set of matching items based on the taxonomy category labels obtained from the hierarchical item taxonomy, wherein the grouping produces a plurality of groups, and wherein each of the plurality of groups comprises items from the set of matching items that are associated with a branch within the hierarchical item taxonomy associated with the category label associated with the group;
• and
• generating a carousel for each of the plurality of groups, wherein a carousel for a group is a scrollable user interface element comprising items associated with the group arranged in a list;
• automatically arrange the generated carousels into a scrollable list based on the stored user data to generate a user interface displaying the set of matching items, wherein the scrollable list is vertically scrollable and wherein automatically arranging the generated carousels comprises:
• applying a machine-learned carousel model to the stored user data, item data describing the set of matching items, and the received search query to generate a score for each of the generated carousels, wherein a generated score for a carousel represents a predicted likelihood that the branch of the hierarchical item taxonomy is relevant to the search query, and
• automatically arranging the generated carousels based on the generated scores, wherein the generated carousels are automatically arranged such that generated carousels with generated scores indicating a higher predicted likelihood are arranged vertically-higher in the scrollable list than generated carousels with generated scores indicating a lower predicted likelihood;
• and
• send instructions for the user interface to the client device for display, wherein sending the instructions for the user interface to the client device causes the client device to automatically display the scrollable list of the plurality of carousels as results to the received search query, wherein the scrollable list of the plurality of carousels are displayed according to the automatic arrangement of the generated carousels, and wherein each of the generated carousels are displayed in the user interface with text describing the taxonomy category label associated with the generated carousel.