**6. Conclusions**

In this chapter, we have reviewed the main concepts related to tags and tag recommendation. There are various sources of data associated with Web 2.0 objects that can be used to extract and rank tags. Candidate tags can be extracted from the textual features associated with the target object using keyword extraction techniques, from mining co-occurrences with other tags, or other textual and content features, and from the neighborhood of the target object and/or target user. We also have briefly discussed various tag quality attributes that can be exploited to rank candidate tags. An effective way to combine these attributes is by means of learn-to-rank techniques, which can automatically "learn" tag recommendation functions from training examples.

### **Acknowledgements**

Our research group is partially funded by Google, the Brazilian National Institute of Science and Technology for Web Research (MCT/CNPq/INCT Web Grant Number 573871/2008-6), and the authors' individual grants from CNPq, CAPES, and FAPEMIG.
