Abstract: Systems and methods of selecting recommendations for a user in an online environment are disclosed. In one aspect, embodiments of the present disclosure include a method, which may be implemented on a system, of performing semantic analysis on a content item associated with the user, online interactions of the user, and profile information related to the user to identify associated content metadata and keywords, assigning a weight to the content metadata and keywords based on semantic-type categories, comparing the content metadata and keywords to target metadata and keywords to identify recommendation matches, and selecting one or more recommendations to be provided to the user based on the recommendation matches.