An analysis of users' behaviour patterns in the organisation of information. A case study of CiteULike
| Pages | 638-656 |
| Published date | 02 August 2013 |
| DOI | https://doi.org/10.1108/OIR-05-2012-0086 |
| Date | 02 August 2013 |
| Author | Ya‐Ning Chen,Hao‐Ren Ke |
| Subject Matter | Information & knowledge management,Library & information science |
An analysis of users’ behaviour
patterns in the organisation of
information
A case study of CiteULike
Ya-Ning Chen and Hao-Ren Ke
Graduate Institute of Library and Information Studies,
National Taiwan Normal University, Taipei, Taiwan
Abstract
Purpose – The purpose of this paper is to investigate the behaviour preferences and patterns of the
organisation of information by taggers, including usage of tags, tag categories and implicit patterns
embedded in social tags.
Design/methodology/approach – The sample was 4,390 social tags (1,777 unique) from 1,661
articles published in 16 library and information science journals selected from CiteULike between
February and March 2011. Using application profiles, a tag category model served as a framework to
develop two sets of hybrid tag categories for analysing the distribution of tag categories and their
implicit patterns.
Findings – The frequency of tag categories was consistent with that of individual tags and obeyed a
power law distribution. In total, six implicit patterns embedded in tags – syntactical, semantic,
mnemonic, genre, contextual hybrid relations and split term – were discovered.
Research limitations/implications – Although this study focused solely on investigating taggers’
behaviour preferences and patterns, the results of this study may shed light on tagging practice, query
formulation and construction of controlled vocabularies.
Originality/value – A set of hybrid tag categories consisting of title, function, content and
topic-related categories is proposed to delineate the distribution of social tags and taggers’ behaviour
preferences, and implicit patterns embedded in tags are generalised. These patterns may be useful for
tagging practice, query formulation and construction of controlled vocabularies.
Keywords Digital libraries,Museums, User studies, Informationmanagement, Social tags,
Information organization, User behaviour,CiteULike
Paper type Research paper
Introduction
Social networking websites such as Flickr and Del.icio.us enable users to organise and
share personal information resources using social tags. Library and information
science (hereafter LIS) experts have tried to understand users’ preferences and patterns
in order to bridge the divide between bottom-up tags created freely by users and the
top-down controlled vocabularies found in professional guidelines for information
organisation. To date several studies in LIS have examined the similarities between
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1468-4527.htm
The work described in this paper was partially supported by the Taiwan E-learning and Digital
Archives Programme sponsored by the National Science Council of Taiwan under NSC Grants
NSC 101-2631-H-001-005, 101-2631-H-001-006 and 101-2631-H-001-014.
OIR
37,4
638
Received 15 May 2012
Second revision approved
12 November 2012
Online Information Review
Vol. 37 No. 4, 2013
pp. 638-656
qEmerald Group Publishing Limited
1468-4527
DOI 10.1108/OIR-05-2012-0086
users’ tags and experts’ controlled vocabularies (Lu et al., 2010; Rolla, 2009; Thomas
et al., 2009; Yi and Chan, 2009). In addition, many studies have proposed categories to
analyse the usage of social tags, and existing categories can be generalised as function
(Angus et al., 2008; Bischoff et al., 2008; Golder and Huberman, 2006; Munk and Mørk,
2007; Sen et al., 2006; Thomas et al., 2009), thesaurus (Kipp, 2011a, b; Spiteri, 2007) and
attribute (Lin et al., 2006; Rorissa, 2010) based categories. Heckner et al. (2008)
developed a tag category model (hereafter TCM), which consisted of linguistic,
functional and tag-2-text sub-models to accommodate all existing categories for usage
analysis of tags, and examined the feasibility of the TCM’s sub-models separately.
Studies have also explored whether taggers would select keywords of information
resources directly as tags (Bischoff et al., 2008; Heymann et al., 2008). There is a need to
combine the three TCM sub-models to obtain an integrated understanding of the
embedded relationships between tags and tagging information resources, such as title-,
content-, function- and topic-related. Some studies have suggested that relationships
embedded in tags could be aggregated into categories which are useful for facilitating
access and improving recall of controlled vocabularies (Rorissa, 2010; Stvilia and
Jo
¨rgensen, 2010). However, to the best of our knowledge no studies have investigated
users’ preferences with a hybrid lens, such as addressing title-, content-, function- and
topic-related categories together, and implicit patterns embedded in social tags to
elucidate personal information organisation behaviour.
Background and problem
With the widespread adoption of social networking websites such as Flickr, Del.icio.us,
Last.fm, YouTube and CiteULike, social tagging or folksonomy has affected not only
users’ organisation of information, but also information organisation in the field of LIS.
Up until now three different approaches have been used to explore the usage of social
tags. First, some studies have attempted to compare tags and title keywords in order to
examine whether taggers select title keywords as tags. Second, several studies have
investigated whether social tags are identical or vary in comparison with the controlled
vocabularies of thesauri. Third, researchers have proposed several categories to
explore the usage of social tags according to specific attributes. Details of these
approaches are given next.
Correspondence between tags and title keywords
Several studies have compared social tags and title keywords. In a study of Del.icio.us
Heymann et al. (2008) compared tags with titles and text of webpages, and found that
half of the tags appeared in the text of webpages and 16 per cent of tags appeared in the
titles of the webpages. Similarly, Bischoff et al. (2008) found that 44.85 per cent of tags
occurred in webpage text, and Heckner et al. (2008) suggested that 26.5 per cent of tags
from Connotea appeared in title keywords of journal articles in the domain of computer
technology. In light of these studies it is interesting to examine whether the
correspondence between tags and title keywords of LIS scholarly journals, where
taggers are potentially equipped with a background in information organisation, is
distinct from those outlined in the existing studies. With reference to the studies
outlined above, this study proposes the following research question:
RQ1. What is the correspondence between social tags (such as CiteULike) and title
keywords of articles in LIS scholarly journals?
A case study of
CiteULike
639
To continue reading
Request your trial