Please follow the link to the jonined workshop

>> K-CAP Workshop on Analyzing Social Media to Represent Collective Knowledge <<

K-CAP 09 Logo -CAP 2009
The Fifth International Conference on Knowledge Capture

First International Workshop on Collective Knowledge Capturing and Representation - CKCaR'09

September 1, 2009
Redondo Beach, California, USA

Workshop Description

The Web 2.0 has introduced a new model of user interactions and encourages people to massively participate in generating and sharing on-line content. This results in a mass amount of data and information available on the Web, created by the collaboration and competition of many individuals. By processing, combining, and integrating these masses of data and information, the Collective Knowledge emerges. Some Web 2.0 platforms already utilize this Collective Knowledge of their users to provide better information and services to their customers. For example, collaborative tagging of several media types can facilitate sharing of such information among users, but also knowledge learned from the analysis of tag networks can provide better annotations for the new incoming media. Collective knowledge is not limited only to social media. Industrial media and corporate or specialized (e.g. biological) knowledge bases can also successfully serve as relevant sources of information, but may require slightly different approaches in extracting new knowledge.

Collective Knowledge is generated in a process that includes analysis and, in many cases, integration of multiple data sources. However, integration of massive or multiple data sources is just a prerequisite for creating Collective Knowledge. As named by Tom Gruber, this is Collected Knowledge (or Collected Intelligence). It becomes Collective Knowledge when new levels of understanding of such knowledge emerge, when "wisdom of the masses" creates new values.

Not all models and representations for Collective Knowledge are created equal. What is easy and natural in one model can be even impossible in the other one. Specific tasks or semantics of information in datasets could require very different approaches and methodologies, producing results of very different quality. Search for collective knowledge always has some aim, like using it to create better book recommendation system or providing better situation understanding for ER teams after earthquake.

The focus of the workshop lies on how new knowledge can be discovered in such datasets, which processes are to be used to harvest the new knowledge, how and in which order the processes are to be applied, and finally, how the resulting Collective Knowledge can be represented. We are specifically interested in the processes that lead to the capturing and creation of this new knowledge and models for representing it. We encourage the presentation and exploration of original work in capturing and representing collective knowledge and expect to stimulate further research in this area. The workshop intends to bring together researchers and practitioners from a wide area of research such as knowledge management, knowledge representation, multimedia analysis, information retrieval, and Semantic Web, who share the common view of the importance of creating, accessing, and using of the collective knowledge. The workshop will enable the participants to discuss the issues and challenges of Collective Knowledge and exchange ideas and experience.

Workshop schedule

The following papers will be presented the joined CKCaR and Social Media Analysis workshop:

    Collective Knowledge Capturing and Representation

  1. 9:00 - 9:30 - Symeon Papadopoulos, Yiannis Kompatsiaris and Athena Vakali. "Leveraging Collective Intelligence through Community Detection in Tag Networks"
  2. 9:30 - 10:00 - Jorge Gracia and Eduardo Mena. "Multiontology Semantic Disambiguation in Unstructured Web Contexts"
  3. 10:00 - 10:30 - Andres Garcia-Silva, Martin Szomszor, Harith Alani and Oscar Corcho. "Preliminary Results in Tag Disambiguation using DBpedia"
  4. 10:30 - 11:00 Coffee break

    Social Media Analysis

  5. 11:00 - 11:30 - Jeon-Hyung Kang, Jihie Kim and Erin Shaw. "Profiling Student Groups in Online Discussion with Network Analysis"
  6. 11:30 - 12:00 - Elena Frantova and Sabine Bergler. "Automatic Emotion Annotation of Dream Diaries"

Workshop Topics

The proposed workshop aims to bring together researchers and practitioners to discuss the current state of the art and open research problems in the knowledge capturing and representation on the massive scale, as well as facilitate collaborations between them. Topics of interest include but are not limited to:

Accepted Papers

The following papers were accepted for the CKCaR workshop:

Paper Submissions

We invite submissions of full technical papers not exceeding 8 pages. Authors of accepted papers will be invited to present their papers in the workshop. Papers must be formatted using the AAAI conference style. Style files for latex and Word can be obtained from the K-CAP 2009 web page ( Submissions must be in PDF format. We will accept submissions via EasyChair system

Submission site at EasyChair (CKCaR'09) is CLOSED.

Important Dates

Submission due EXTENDED: June 15 25, 2009 - CLOSED

Notification of acceptance: July 28, 2009

Camera-ready versions due: August 16, 2009

Workshop: September 1, 2009


Maciej Janik, University of Koblenz-Landau, Germany, janik[at] (main contact)

Ansgar Scherp, University of Koblenz-Landau, Germany, scherp[at]

Yiannis Kompatsiaris, ITI, Thessaloniki, Greece, ikom[at]

Program Committee

Mathieu d'Aquin, Open University, UK

Christopher Brewster, Aston University, UK

Oren Etzioni, University of Washington, USA

Olivier Gerbe, HEC Montreal, Canada

Bettina Hoser, University of Karlsruhe, Germany

Marta Sabou, Open University, UK

Vojtech Svatek, University of Economics, Prague, Czech Republic

Dan G. Tecuci, University of Texas at Austin, USA