Dissertation Information for Xin Xiang NAME: - Xin Xiang
DEGREE:
- Ph.D.
DISCIPLINE:
- Library and Information Science
SCHOOL:
- University of Illinois, Urbana-Champaign (USA) (2011)
ADVISORS: - Linda C. Smith
COMMITTEE MEMBERS: - P. Bryan Heidorn - John Unsworth - Chen Xiang Zhai
MPACT Status: Fully Complete
Title: A network approach to topic summary and knowledge discovery in social tagging
Abstract: As evidenced by the growing popularity of collaborative tagging sites like librarything , last.fm and del.icio.us , social tagging has provided a social and information organizing platform that warrants public attention and academic investigation alike. This doctoral research focuses on studying the semantic relations between social tags, items and content creators through co-occurrence analysis, social network analysis and information visualization, thus revealing the role played by social tags in representing and classifying contents and creators, and implications they might have for facilitating information seeking practice, particularly knowledge discovery and information summary, and as a result, helping the design of information retrieval and browsing interface. User-oriented studies are conducted to evaluate the advantage of visual and presentational features based on tagging analysis over existing constructs such as tag clouds in performing high-level information seeking tasks.
The social tagging paradigm is widely considered an extension beyond keyword-based indexing and hierarchical classification schemes. The new massive manual indexing method characterized by social tagging differs from automatic indexing that lays the foundation of modern information retrieval in that its manual nature obviates the common pitfalls of computer-based automatic indexing. It also complements traditional manual indexing since tag word distribution reflects the opinions of a large number of people with various background and knowledge instead of a limited number of domain experts who are dominant in the classification and cataloging undertakings.
Parallel to the observation that an individual's social identity is defined by the collectivities to which the individual belongs, the topical, temporal, geographic, and stylistic features of an information item (book, song, etc.) are represented by the tags that are applied to it in a social tagging context. Employing similarity analysis, bipartite social network theory and small-world network model, this study analyzes the patterns and trends of networks formed through co-occurrences of tags, and clusters and community structures found in the networks that convey topical or stylistic cues of the underlying items.
The abundant tagging data available at public tagging sites makes it possible to reveal the relations between tags, items and creators from a social network perspective. A small-world network, characterized by low average path length and high clustering coefficient compared to a randomly generated network with a similar number of nodes and edges, is a typical form of social network frequently found in real world social networks and physical networks. The study demonstrates the small-world network property of networks of tags/authors in the book tagging site librarything , presenting the network of tags/authors as groups of highly related items which can be detected by community detection methods and convey semantic meanings.
The big picture of the tagging universe aside, several network reduction techniques are used to contract the original tag/author networks containing thousands of nodes to a smaller network of around 150 nodes for better visualization and presentation of details, especially of those most salient tags and creators. The exact number of nodes in the contracted network is fine-tuned based on weights of edges and topological characteristics of the network. Networks of different scales seem to present different levels of information about the tagging world.
A user study is conducted to evaluate the effectiveness of visual constructions based on similarity and network analysis for several tasks that they can be used to support. Participants are presented with both visualizations based on tagging analysis and an existing interface such as tag cloud, and asked to determine which one is more instructive and helpful in performing different high-level information seeking tasks, such as topic summary, grouping and navigation.
This dissertation work aims at studying tagging data from a social network perspective, providing clues as to how analysis of tagging data offers new angles to interpret semantic relations and facilitate content presentation and discovery. It seeks to strengthen the function of social tagging as a point of connection between personal content management and serendipitous knowledge discovery in a social context.
| |
MPACT Scores for Xin Xiang A = 0
C = 0
A+C = 0
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2012-07-30 12:06:08
Advisors and Advisees Graph
generating graph, please reload |