(In) formal classification : connecting formal and informal knowledge organization systems
MetadataVis full innførsel
Recent developments in knowledge organization systems (KOS) have added new dimensions to traditional classification practices, as knowledge is now not only organized by official committees and editors, but also by the users of these systems themselves. The aim of this study is to provide new insights into mapping formally created knowledge organization systems to socially created knowledge organization systems. To this end, a literature review, a theoretical analysis and a practical analysis have been carried out. The theoretical analysis consists of a comparison between the knowledge structure of Freebase, a semantic encyclopedia based on Linked Open Data, and the Dewey Decimal Classification (DDC), a traditional classification system. The practical analysis is based on a statistical mapping of Freebase “domains” to classes of the Dewey Decimal Classification, using information available in the WorldCat database. This analysis is of a quantitative nature. The findings of this study can be approached on multiple levels. On a broad level, it shows that it is possible to map the structure of a socially created knowledge organization system to the structure of a traditional, formally created system, although the study also indicates that there are some fundamental differences in these systems that cannot be solved in an easy way. When we look at the level of the statistical mapping between Freebase and the DDC, this study shows that the mapping covers most of the available Freebase domains, and that it could feasible to use this approach on a broader scale. On the other hand, the study also shows that there are limits as to which features of Freebase's knowledge structure can be represented using the DDC. Structure This thesis starts with an outline of the topic and research questions, and continues with an extensive literature review, that discusses formally and socially created KOS, and theories and practices in mapping. Subsequently, the methodology is discussed, followed by a theoretical analysis of the structure of the DDC and Freebase, and a practical analysis based on a statistical mapping. Finally, the thesis is wrapped up in the conclusion, that also indicates suggestions for future research.
Joint Master Degree in Digital Library Learning (DILL)
UtgiverHøgskolen i Oslo og Akershus. Institutt for arkiv, bibliotek- og info.fag
Universitetet i Tallinn
Universitetet i Parma