Generative artificial intelligence and preference aware hashtag generation

An electronic device comprising:

• at least one processing device configured to:
• retrieve content information related to content from a content source;
• input the content information into a machine learning model, wherein the machine learning model is trained to recognize features of the content information using a combination of at least two of video, audio, and metadata features extracted from the content;
• receive a plurality of predicted hashtags for the content from the machine learning model, wherein the plurality of predicted hashtags is generated by the machine learning model using the input content information, wherein each one of the plurality of predicted hashtags includes a probability score;
• score the plurality of predicted hashtags based on the probability score of each one of the plurality of predicted hashtags;
• retrieve hashtag preference information including one or more hashtags from a hashtag history database associated with a user device of a user;
• determine one or more preference features associated with the one or more hashtags using the hashtag preference information;
• re-score and sort the plurality of predicted hashtags using the one or more preference features;
• and
• output at least one of the plurality of predicted hashtags.

A non-transitory machine readable medium containing instructions that when executed cause at least one processor of an electronic device to:

• retrieve content information related to content from a content source;
• input the content information into a machine learning model, wherein the machine learning model is trained to recognize features of the content information using a combination of at least two of video, audio, and metadata features extracted from the content;
• receive a plurality of predicted hashtags for the content from the machine learning model, wherein the plurality of predicted hashtags is generated by the machine learning model using the input content information, wherein each one of the plurality of predicted hashtags includes a probability score;
• score the plurality of predicted hashtags based on the probability score of each one of the plurality of predicted hashtags;
• retrieve hashtag preference information including one or more hashtags from a hashtag history database associated with a user device of a user;
• determine one or more preference features associated with the one or more hashtags using the hashtag preference information;
• re-score and sort the plurality of predicted hashtags using the one or more preference features;
• and
• output at least one of the plurality of predicted hashtags.