TY - JOUR
T1 - Exploring the structure of library and information science web space based on multivariate analysis of social tags
AU - Joo, Soohyung
AU - Kipp, Margaret E.I.
N1 - Publisher Copyright:
© the author, 2015.
PY - 2015
Y1 - 2015
N2 - Introduction. This study examines the structure of Web space in the field of library and information science using multivariate analysis of social tags from the Website, Delicious.com. A few studies have examined mathematical modelling of tags, mainly examining tagging in terms of tri-partite graphs, pattern tracing and descriptive statistics. This study is one of the few studies to employ multivariate analysis in investigating dimensions of Web spaces based on social tagging data. Method. This study examines the post data collected from a set of library and information science related Websites bookmarked on Delicious.com using a Web crawler. Post data consist of the URL, usernames, tags and comments assigned by users of Delicious.com. The collected tag data were analysed based on multivariate methods, such as multidimensional scaling and structural equation modelling. Analysis. Collected data were first analysed using multidimensional scaling to explore initial relationships amongst the selected Websites. Then, confirmatory factor analysis based on structural equation modelling was employed to examine the hierarchical structure of the library & information science Web space. Results. Social tag data exhibit different dimensions in the Web space of the library and information science field. In addition, social tags confirmed the hierarchical structure of the field by showing significantly stronger relationships between the sites with similar characteristics. That is, the structure of the tagging data shows similar connections to those present in the real world. Conclusions. This study suggests a new statistical approach in social tagging and Web space analysis studies. Tag information can be used to explain the hierarchical structure of a certain domain. Methodologically, this study suggests that structural equation modelling can be a compelling method to explore hierarchal structures of nodes on the Web space.
AB - Introduction. This study examines the structure of Web space in the field of library and information science using multivariate analysis of social tags from the Website, Delicious.com. A few studies have examined mathematical modelling of tags, mainly examining tagging in terms of tri-partite graphs, pattern tracing and descriptive statistics. This study is one of the few studies to employ multivariate analysis in investigating dimensions of Web spaces based on social tagging data. Method. This study examines the post data collected from a set of library and information science related Websites bookmarked on Delicious.com using a Web crawler. Post data consist of the URL, usernames, tags and comments assigned by users of Delicious.com. The collected tag data were analysed based on multivariate methods, such as multidimensional scaling and structural equation modelling. Analysis. Collected data were first analysed using multidimensional scaling to explore initial relationships amongst the selected Websites. Then, confirmatory factor analysis based on structural equation modelling was employed to examine the hierarchical structure of the library & information science Web space. Results. Social tag data exhibit different dimensions in the Web space of the library and information science field. In addition, social tags confirmed the hierarchical structure of the field by showing significantly stronger relationships between the sites with similar characteristics. That is, the structure of the tagging data shows similar connections to those present in the real world. Conclusions. This study suggests a new statistical approach in social tagging and Web space analysis studies. Tag information can be used to explain the hierarchical structure of a certain domain. Methodologically, this study suggests that structural equation modelling can be a compelling method to explore hierarchal structures of nodes on the Web space.
UR - http://www.scopus.com/inward/record.url?scp=85000632572&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85000632572&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:85000632572
SN - 1368-1613
VL - 20
JO - Information Research
JF - Information Research
IS - 4
M1 - 696
ER -