| Summary: | Social Classification or tagging is increasingly a subject of interest in library and information science (and related fields) as social bookmarking tools such as del.icio.us have become increasingly popular. Simple visualisations such as sorting tags by frequency or displaying tag clouds in which tag size denotes popularity suggest that tagging systems form interesting new taxonomies or folksonomies of related terms. This study examines these tagging systems seeking elements of convergence and divergence with traditional cataloguing and classification theory and practice.
This study examines the use of unusual tags which do not fit the mould of traditional cataloguing and classification, namely, that they are not subject related. These tags include two major categories:
affective (emotional) tags,
time and task related tags.
Examples of affective tags include interesting, fun and cool. Examples of time and task related tags include @toread, todo, and tobuy.
Data has been collected from del.icio.us, citeulike and connotea via python scripts designed to gather information on specific tags or URLs. The data used for this study is part of a larger study by Kipp and Campbell (2006) examining patterns in tagging. Analysis of this data showed approximately 16% of tags were time and task related. Time and task or affective tags were located in multidimensional scaling graphs of cotag (coword) data. (Kipp and Campbell 2006)
Many users of del.icio.us, citeulike and connotea appear to want to store more than just the subject of the documents they are bookmarking. Tags such as @toread, tobuy, todo, fun and cool suggest that users see their relationship to these documents in different ways. While the latter tags express an emotional connection to the document, the former show evidence of a desire to attach personal information management information to documents. This desire to combine personal information management and document classification echoes findings in document use research at Xerox in which users categorised items in order to better understand their relationship to other items and to tasks the users wished to perform. (Sellen and Harper 2002)
A large part of library science research is involved in the examination of how users seek and use information. Another important aspect of this is how they relate to information. (Bates 1998, 1048) Findings from this study suggest that users relate information to time related tasks, activities and their own emotional reactions.
This poster was presented at 17th Annual SIG/CR Classification Research Workshop part of the Annual Meeting of the American Society for Information Science and Technology (ASIST), November 4, 2006, Austin, Texas
|
|---|
