Search Query Generation Based Upon Received Text

A computing device comprising:

• a processor;
• and
• memory comprising processor-executable instructions that when executed by the processor cause performance of operations, the operations comprising:
• determining a similarity profile indicative of similarity scores associated with similarities between terms of a set of terms;
• determining, based upon the similarity profile, relevance scores associated with the set of terms;
• selecting, based upon the relevance scores, one or more search terms from among the set of terms;
• performing a search based upon the one or more search terms;
• sending, to a client device, a set of search results of the search;
• receiving, from the client device, a selection of an image of a plurality of images in the set of search results;
• responsive to the selection of the image, combining at least a portion of the image with first set of text to generate a content item;
• and
• at least one of presenting the generated content item to one or more other client devices or determining a target audience for the generated content item.

A non-transitory machine readable medium having stored thereon processor-executable instructions that when executed cause performance of operations, the operations comprising:

• displaying a content item creation interface via a client device;
• receiving, via the content item creation interface, a first set of content item text;
• selecting, based upon the first set of content item text and a plurality of sets of content item text associated with a plurality of content items, a set of content items from among the plurality of content items;
• determining, based upon the first set of content item text and the set of content items, a set of terms;
• determining embedding-based representations, of the set of terms, comprising:
• a first embedding-based representation of a first term of the set of terms;
• a second embedding-based representation of a second term of the set of terms;
• and
• a third embedding-based representation of a third term of the set of terms;
• determining a first measure of similarity between the first embedding-based representation and the second embedding-based representation;
• determining a second measure of similarity between the first embedding-based representation and the third embedding-based representation;
• and
• determining, based upon the first measure of similarity and the second measure of similarity, a similarity profile indicative of similarity scores associated with similarities between terms of the set of terms.