Sung-Kwon Yang

Social semantic cloud of tag: semantic model for social tagging

Tagging has proven to be a successful and efficient way for creating metadata through a human collective intelligence. It can be considered not only an application of individuals for expressing ones interests, but also as a starting point for leveraging social connections through collaborative user participations. A number of users have contributed to tag resources in web sites such as Del.icio.us, Flickr etc. However, there is no uniform structure to describe tags and users activities. This makes difficult to share and represent tag data among people. The SCOT (Social Semantic Cloud of Tags) ontology is aimed to represent the structure and semantics of a set of tags and promotes their global sharing. The paper introduce the SCOT ontology and methods of its representation.

Simple Algorithms for Representing Tag Frequencies in the SCOT Exporter

In this paper we describe the SCOT Exporter and its algorithms to create instance data based on the SCOT (Social Semantic Cloud of Tags) ontology for sharing and reusing tag data. The algorithms use tag frequencies and co-occurrence relations to represent statistical information via the SCOT ontology. We give an overview of the Exporter and the algorithms, and then discuss some experimental results.

Dynamic generation of concepts hierarchies for knowledge discovering in bio-medical linked data sets

Since most bio-medical Linked Data Sets are simply extracted from the Relational database, lots of them are lack of ontology or concept hierarchy structure for user better understanding the data sets. This problem also limited usage of bio-medical Linked Data Sets. To resolve the problem, this paper introduced a method to dynamically generate the concept hierarchy from the Linked Data Sets. Based on the hierarchical clustering algorithm, we applied Vector Space Model(VSM) and Jaccards Coefficient(JC) to formalize the hierarchy structure after pre-processing data. We implemented our method using two Linked Data Sets: DrugBank and Diseasome from Linked Life Data and evaluated performance with the gold standard.

Implementing Bibliographic Metadata Model for Social Semantic Digital Libraries

We propose semantic model that is possible to apply for the bibliographic metadata of domestic digital library by analysing bibliographic metadata models like MARC, DC, MODS, JeromeDL`s metadata model MarcOnt as the representative case of semantic digital library and FRBR model as the conceptual model.

Browsing Unbounded Social, Linked Data Instances with User Perspectives

With the emerging uses of semantically enriched social data on the Web, linked data are expected to envision a next generation of the current web. As ‘web of data’, they are spread as pieces of data into the Web with links to related objects or concepts. Data instances distributed with URIs, those that enable identification and combination of data instances, can be consumed with shared data vocabularies. Easy and intuitive access to the data should be provided for data-centered uses of the Web. This paper introduces a linked data browser providing an intuitive view, especially helping casual users’ understanding of data instances and their relationships. The browser also satisfies the requirement of a generic browser: handling unexpected domains of data across the links. By adapting user’s perspectives captured during browsing, the browser enables users to view any types of linked data instances with different views pertinent to their intentions and types of data.

STAN: Social, Trusted Annotation Network

Annotated data play an important role in enhancing the usability of information resources. Single users can be easily frustrated by the task of annotating. Collaborative approaches to annotation have been applied to web resources, but have not yet been applied to the task of local documents, due in part to the lack of a uniform identification method. In this paper, we use hash-based virtual URIs for identifying documents, and introduce the concept of a STAN (Social, Trusted Annotation Network), which enables collaborative annotation of documents through their URIs. STAN also incorporates quantitative trust rates between users in social networks based on their interactions with each other. The STAN framework is described, demonstrating how these trust networks are constructed through collaborative annotation. Finally, we evaluate the usefulness of collaborative annotation and the feasibility of the resulting trust rates through empirical experiment.