Abstract
The development of social tagging systems on the Web under the participatory movement often called "Web 2.0", has enabled the emergence of new forms of indexing web content, allowing users to categorize their resources by associating their keywords, called tags. Considered as tripartite hypergraphs of users, tags and resources, the resulting structures, aka folksonomies, are of great interest in information retrieval. In this paper, we exploit these triples (resources, users, tags) to introduce a new definition of a generic basis of triadic association rules, called BGRT. We show that the use of generic rules for autocompletion feature highlights the relevance of folksonomies and their real interest in the case of information retrieval. The first results obtained on a real folksonomy are promising and offer many opportunities.
| Original language | French |
|---|---|
| Journal | Document Numerique |
| Volume | 15 |
| Issue number | 1 |
| Pages (from-to) | 101-124 |
| ISSN | 1279-5127 |
| DOIs | |
| Publication status | Published - 2012 |
| Externally published | Yes |
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver