Evaluation of Folksonomy Induction Algorithms

摘要: Algorithms for constructing hierarchical structures from user-generated metadata have caught the interest of academic community in recent years. In social tagging systems, output these algorithms is usually referred to as folksonomies (from folk-generated taxonomies). Evaluation and folksonomy induction a challenging issue complicated by lack golden standards, comprehensive methods tools well research empirical/simulation studies applying methods. this article, we report results broad comparative study state-of-the-art that applied evaluated context five systems. addition adopting semantic evaluation techniques, present adopt new technique can be used evaluate usefulness navigation. Our work sheds light on properties characteristics introduces pragmatic approach evaluation, while at same time identifying some important limitations challenges evaluation. show specifically developed capture intuitions systems outperform traditional clustering techniques. To best our knowledge, represents largest most date.