Associating a graphical element to media content item collections

Abstract

Various embodiments provide for associating a collection of media items with a graphical element. For instance, a system can: generate corpus data from a set of features of a collection of media content items; determine a set of candidate graphical elements for the collection of media content items based on the corpus data and further based on a set of first mappings associating at least one graphical element and at least one n-gram; determine a set of prediction scores corresponding to the set of candidate graphical elements based on the set of features; determine a ranking for the set of candidate graphical elements based on the set of prediction stores; select a set of predicted graphical elements, from the set of candidate graphical elements, based on the ranking; and provide the set of predicted graphical elements in association with the collection of media content items.

What is claimed is: 
     
         1 . A method comprising:
 determining, by one or more processors, a set of candidate graphical elements for a collection of media content items based on a set of textual n-grams, based on a set of first mappings that maps at least one textual n-gram in the set of textual n-grams to a first graphical element, and based on a set of second mappings that maps at least the first graphical element to a second graphical element, the set of textual n-grams being identified in a set of features of the collection of media content items, the second graphical element being determined to be one candidate graphical element in the set of candidate graphical elements;   determining, by the one or more processors, a set of prediction scores corresponding to the set of candidate graphical elements based on the set of features of the collection of media content items;   selecting, by the one or more processors, a set of predicted graphical elements, from the set of candidate graphical elements, based on the set of prediction scores; and   providing, by the one or more processors, the set of predicted graphical elements in association with the collection of media content items.   
     
     
         2 . The method of  claim 1 , wherein the set of candidate graphical elements includes at least one of an emoticon or an emoji. 
     
     
         3 . The method of  claim 1 , further comprising:
 tuning, by the one or more processors, the determining of the set of prediction scores based on the set of features by adjusting at least one weight used in calculating a prediction score for at least one graphical element in the set of candidate graphical elements.   
     
     
         4 . The method of  claim 1 , wherein the set of features further comprises at least one of a caption or a particular graphical element associated with the collection of media content items. 
     
     
         5 . The method of  claim 1 , further comprising:
 generating, by the one or more processors, the set of first mappings that maps the at least one textual n-gram in the set of textual n-grams to a first graphical element based on data that provides a Unicode standard description for the first graphical element.   
     
     
         6 . The method of  claim 1 , further comprising:
 generating, by the one or more processors, the set of first mappings that maps the at least one textual n-gram in the set of textual n-grams to a first graphical element based on co-occurrences of the first graphical element and the at least one textual n-gram with respect to at least one other collection of media content items.   
     
     
         7 . The method of  claim 1 , wherein the set of second mappings maps two or more graphical elements together. 
     
     
         8 . The method of  claim 7 , further comprising:
 generating, by the one or more processors, the set of second mappings mapping two or more graphical elements together based on the set of first mappings that maps the at least one textual n-gram in the set of textual n-grams to a first graphical element.   
     
     
         9 . A system comprising:
 one or more processors; and   one or more machine-readable mediums storing instructions that, when executed by the one or more processors, cause the system to perform operations comprising:
 determining a set of candidate graphical elements for a collection of media content items based on a set of textual n-grams, based on a set of first mappings that maps at least one textual n-gram in the set of textual n-grams to a first graphical element, and based on a set of second mappings that maps at least the first graphical element to a second graphical element, the set of textual n-grams being identified in a set of features of the collection of media content items, the second graphical element being determined to be one candidate graphical element in the set of candidate graphical elements; 
 determining a set of prediction scores corresponding to the set of candidate graphical elements based on the set of features of the collection of media content items; 
 selecting a set of predicted graphical elements, from the set of candidate graphical elements, based on the set of prediction scores; and 
 providing the set of predicted graphical elements in association with the collection of media content items. 
   
     
     
         10 . The system of  claim 9 , wherein the set of candidate graphical elements includes at least one of an emoticon or an emoji. 
     
     
         11 . The system of  claim 9 , wherein the operations further comprise:
 tuning the determining of the set of prediction scores based on the set of features by adjusting at least one weight used in calculating a prediction score for at least one graphical element in the set of candidate graphical elements.   
     
     
         12 . The system of  claim 9 , wherein the set of features further comprises at least one of a caption or a particular graphical element associated with the collection of media content items. 
     
     
         13 . The system of  claim 9 , wherein the operations further comprise:
 generating the set of first mappings that maps the at least one textual n-gram in the set of textual n-grams to a first graphical element based on data that provides a Unicode standard description for the first graphical element.   
     
     
         14 . The system of  claim 9 , wherein the operations further comprise:
 generating the set of first mappings that maps the at least one textual n-gram in the set of textual n-grams to a first graphical element based on co-occurrences of the first graphical element and the at least one textual n-gram with respect to at least one other collection of media content items.   
     
     
         15 . The system of  claim 9 , wherein the set of second mappings maps two or more graphical elements together. 
     
     
         16 . The system of  claim 15 , wherein the operations further comprise:
 generating the set of second mappings mapping two or more graphical elements together based on the set of first mappings that maps the at least one textual n-gram in the set of textual n-grams to a first graphical element.   
     
     
         17 . A non-transitory computer-readable medium storing instructions that, when executed by one or more computer processors, cause the one or more computer processors to perform operations comprising:
 determining a set of candidate graphical elements for a collection of media content items based on a set of textual n-grams, based on a set of first mappings that maps at least one textual n-gram in the set of textual n-grams to a first graphical element, and based on a set of second mappings that maps at least the first graphical element to a second graphical element, the set of textual n-grams being identified in a set of features of the collection of media content items, the second graphical element being determined to be one candidate graphical element in the set of candidate graphical elements;   determining a set of prediction scores corresponding to the set of candidate graphical elements based on the set of features of the collection of media content items;   selecting a set of predicted graphical elements, from the set of candidate graphical elements, based on the set of prediction scores; and   providing the set of predicted graphical elements in association with the collection of media content items.   
     
     
         18 . The non-transitory computer-readable medium of  claim 17 , wherein the set of candidate graphical elements includes at least one of an emoticon or an emoji. 
     
     
         19 . The non-transitory computer-readable medium of  claim 17 , wherein the operations further comprise:
 tuning the determining of the set of prediction scores based on the set of features by adjusting at least one weight used in calculating a prediction score for at least one graphical element in the set of candidate graphical elements.   
     
     
         20 . The non-transitory computer-readable medium of  claim 17 , wherein the set of features further comprises at least one of a caption or a particular graphical element associated with the collection of media content items.