Abstract
Social-bookmarking systems such as del.icio.us let users label resources with freely chosen key- words that they find most useful for their intentions. These keywords are also known as tags. The cumulated information from tag assignments by a large community of users constitutes a folksonomy. This thesis analyses how well tag assignments in a folksonomy, that are assigned to semistructured English texts can be found by automatic text classification systems. Therefore, the machine learning algorithms SVM, k-NN, multinomial naive Bayes, and Rocchio have been evaluated for their effectiveness with such data.
Users
Please
log in to take part in the discussion (add own reviews or comments).