Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for classifying media accounts of a media system into different demographic buckets based on media content data associated with the media accounts, where media content associated with a media account can be accessed by a plurality of members in a household of the media account. A demographic classifier can receive a plurality of media accounts, a plurality of sets of probability scores corresponding to the plurality of media accounts, and a plurality of demographic buckets. The demographic classifier can further estimate, based on public available information, an approximate number of media accounts of the plurality of media accounts belonging to a specific demographic bucket. The demographic classifier can further select the approximate number of media accounts to be included in a set of media accounts assigned to the specific demographic bucket.