A method, computer system, and computer program product for collaborative document relations modeling are provided. The embodiment may include parsing, by a processor, content of a web conference recording document utilizing multi-modal analysis. The embodiment may also include deriving a social meeting bookmark model based on the multi-modal analysis to combine results of the multi-modal analysis into a matrix. The embodiment may further include generating bookmarks of the parsed content based on the derived social meeting bookmark model and the matrix. The embodiment may also include storing the bookmarks and the social meeting bookmark model in a database.