• Journal of the Korean Society for Library and Information Science
• /
• v.44 no.4
• /
• pp.203-230
• /
• 2010

This study analyzed the development and growth of folksonomy by examining tags associated with 40 bestsellers on LibraryThing.com in 6-month intervals. It was found that tag values do not decrease but grow in terms of quantity and quality. Accordingly, we examined the major significances of the tags and their potential utilization as an expression of subjects. Our findings were as follows. First, the motivations for tagging can be categorized into personal information for search purposes, self-fulfillment such as sense of achievement, display of emotion and sharing of one's experience with others, or an altruistic objective that emphasizes sociality with a desire that one's actions might provide social benefits. According to our analysis, 74.12% of tags had a social motivation. Second, the total number of tags and the frequency of usage increased with time. Third, the categories that showed a high increase in tag usage were dates of publication and reading, key words, main characters, and book reviews. Tags related to subjects had the highest ratio. Fourth, among Library of Congress Subject Headings (LCSH), multiple genres, key words and main characters were assigned to books, and specific key words and other properties were added as time progressed. There was also a slight increase in the number of tags consistent with LCSH. Fifth, we found that key tags could serve as a compilation of terms that reflects the knowledge base of the corresponding era. Thus, folksonomy should be continuously monitored for its quantitative and qualitative development of the tags to make improvements on its formative disadvantages, and identify internal semantic significance, be actively utilized in conjunction with taxonomy as a flexible compilation of terms that incorporate the history of a specific era.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.7
• /
• pp.221-228
• /
• 2020

Part-of-Speech (POS) tagged corpus is a collection of electronic text in which each word is annotated with a tag as the corresponding POS and is widely used for various training data for natural language processing. The training data generally assumes that there are no errors, but in reality they include various types of errors, which cause performance degradation of systems trained using the data. To alleviate this problem, we propose a novel method for detecting errors in the existing POS tagged corpus using the classifier of XGBoost and cross-validation as evaluation techniques. We first train a classifier of a POS tagger using the POS-tagged corpus with some errors and then detect errors from the POS-tagged corpus using cross-validation, but the classifier cannot detect errors because there is no training data for detecting POS tagged errors. We thus detect errors by comparing the outputs (probabilities of POS) of the classifier, adjusting hyperparameters. The hyperparameters is estimated by a small scale error-tagged corpus, in which text is sampled from a POS-tagged corpus and which is marked up POS errors by experts. In this paper, we use recall and precision as evaluation metrics which are widely used in information retrieval. We have shown that the proposed method is valid by comparing two distributions of the sample (the error-tagged corpus) and the population (the POS-tagged corpus) because all detected errors cannot be checked. In the near future, we will apply the proposed method to a dependency tree-tagged corpus and a semantic role tagged corpus.