Can all tags be used for search?
Proceedings of the 17th ACM conference on Information and knowledge management
Latent dirichlet allocation for tag recommendation
Proceedings of the third ACM conference on Recommender systems
Language Models and Topic Models for Personalizing Tag Recommendation
WI-IAT '10 Proceedings of the 2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology - Volume 01
Detect'11: international workshop on DETecting and Exploiting Cultural diversiTy on the social web
Proceedings of the 20th ACM international conference on Information and knowledge management
| Hi-index | 0.00 |
The social Web has led to a flood of user-generated content in the last years being both a blessing and a curse. On the one hand, the individual user is totally overwhelmed by the endless stream of data and ubiquitous information. On the other hand, this huge amount of data can be explored and analyzed automatically to enable accurate predictions and behavioral studies. We will look into a particular social web application: tagging systems annd investigate how personal and diverse tag recommendations can be realized exploiting the wisdom of the crowd. Tagging systems provide a platform for users to share, retrieve, and organize their social annotations. These tags fulfill different purposes and can be used in different ways. In the first part, we will investigate the different tag types and how they can be exploited for different task. Based on the work by Bischoff et al. [1] we look at a classification scheme for social annotations and analyse different tagging behavior. In the second part, we will look into tag recommendation and how we can achieve good results with respect to the whole system and with respect to individual users. Collective tag recommendation using topic modeling techniques [3] and the combination with a language modeling approach [2] will be presented. Besides balancing diversity and personal preferences, we will also look into evaluation methods for tag recommendation and discuss possible problems and dependencies.