Mining tags using social endorsement networks

Entities on social systems, such as users on Twitter, and images on Flickr, are at the core of many interesting applications: they can be ranked in search results, recommended to users, or used in contextual advertising. Such applications assume knowledge of an entity's nature and characteristic attributes. An effective way to encode such knowledge is in the form of tags. An untagged entity is practically inaccessible, since it is hard to retrieve or interact with. To address this, some platforms allow users to manually tag entities. However,while such tags can be informative, they can oftentimes be inadequate, trivial, ambiguous, or even plain false. Numerous automated tagging methods have been proposed to address these issues. However,most of them require pre-existing high-quality tags or descriptive texts for every entity that needs to be tagged. In our work, we propose a method based on social endorsements that is free from such constraints. Virtually every major social networking platform allows users to endorse entities that they find appealing. Examples include "following" Twitter users or "favoriting" Flickr photos. These endorsements are abundant and directly capture the preferences of users. In this paper, we pose and solve the problem of using the underlying social endorsement network to extract useful tags for entities in a social system. Our work leverages techniques from topic modeling to capture the interests of users and then uses them to extract relevant and descriptive tags for the entities they endorse. We perform an extensive evaluation of our proposed approach on real large-scale datasets from both Twitter and Flickr, and show that it significantly outperforms meaningful and competitive baselines.