An Intelligent CSCW Workbench: Analysis, Visualization, and Agents

Principal investigator:
Dr. Hsinchun Chen

Department of Management Information Systems, Artificial Intelligence Lab

The University of Arizona

Contact Information

Dr. Hsinchun Chen

Department of Management Information Systems, Artificial Intelligence Lab
The University of Arizona

McClelland Hall Ste. 430
1130 E. Helen St.
Tucson, AZ 85721
Phone: (520) 621- 2748
Fax : (520) 621-2433
Email: hchen@bpa.arizona.edu

URL: http://ai.bpa.arizona.edu/hchen/

WWW PAGE

http://ai.bpa.arizona.edu

List of Supported Students and Staff

Bin Zhu, doctoral candidate

Roslin Hauck, doctoral candidate

Haiyan Fan, masters candidate

Project Award Information

· Award Number: IIS-9800696

· Duration: June 1998-June 2001

· Title: An Intelligent CSCW Workbench: Analysis, Visualization, and Agents

Keywords

CSCW, Information Analysis, Information Visualization, Agents, Organizational Computing

Project Summary

We are developing an intelligent agent-based CSCW workbench to explore research issues related to coordinated work. Information analysis, information visualization, and agent techniques are being used to address such CSCW problems as: idea convergence difficulty, information overload, vocabulary difference, increased cognitive demand, and interface/display problems. This workbench allows mixing and matching of different analysis, visualization, and agent components. Evaluation of the feasibility and effectiveness of these “technological cocktail mixes” are currently being performed.

Publications and Products

Journals

K. Tolle and H. Chen, “Comparing Noun Phrasing Techniques for Use with Medical Digital Library Tools,” Journal of the American Society for Information Science, Special Issue on Digital Libraries, Volume 51, Number 4, Pages 352-370, 2000.

H. Chen, “Introduction to the Special Topic Issue: Part 1, Digital Libraries, The Field,” Journal of the American Society for Information Science, Special Issue on Digital Libraries, Volume 51, Number 3, Pages 213-215, 2000.

H. Chen, “Introduction to the Special Topic Issue: Part 2, Towards Building a Global Digital Library,” Journal of the American Society for Information Science, Special Issue on Digital Libraries, Volume 51, Number 4, Pages 311-312, 2000.

B. Zhu and H. Chen, “Validating a Geographic Image Retrieval System,” Journal of the American Society for Information Science, Volume 51, Number 7, Pages 625-634, 2000.

D. G. Roussinov and H. Chen, “Information Navigation on the Web by Clustering and Summarizing Query Results,” Information Processing and Management, 2000, forthcoming.

G. Leroy and H. Chen, "Meeting medical terminology needs: the ontology-enhanced Medical Concept Mapper," IEEE Transactions on Information Technology in Biomedicine, forthcoming 2001.

H. Chen, H. Fan M. Chau, and D. Zeng, “MetaSpider: Meta-Searching and Categorization on the Web.” Journal of the American Society for Information Science and Technology, forthcoming, 2001.

N. C. Romano, C. Bauer, H. Chen and J. F. Nunamaker, “Quantifying Qualitative Data for Electronic Commerce Attitude Assessment and Visualization,” Journal of Management Information Systems, 2000, forthcoming.

Other specific products

CI Spider and Meta Spider for Personalized Web Analysis: The two agent-based systems support information collection from multiple web data sources and dynamic, interactive information categorization and visualization for web searchers. The systems are available for free download at http://ai.bpa.arizona.edu. More than 1000 users have downloaded our software since November 1999.

CI Spider:

The user inputs the starting URL and search phrase into the CI Spider. Search options such as maximum number of pages to be retrieved and boolean search options are specified at this point. The search results are displayed dynamically. The Good URL List shows all the web pages containing the search phrase. Noun phrases are extracted from the web pages and the user can select preferred phrases for categorization. Finally, the SOM is generated based on the phrases selected. Steps 3 or 4 can be done iteratively to refine the results.

Meta Spider:

The user chooses the search engines to be include and enters the search query into Meta Spider. Search options such as maximum number of pages to be retrieved and boolean search options are specified at this point. Search results from chosen search engines are displayed dynamically. Web pages are fetched from the Internet for further analysis. Noun phrases are extracted from the web pages and the user can select preferred phrases for categorization. Finally, the SOM is generated based on the phrases selected. Steps 3 or 4 can be done iteratively to refine the results.

Project Impact

Human Resources:

Doctoral students involved in the various experiments sponsored by this project received the opportunity to design and conduct experiments. Literature reviews, system implementation, research methodology considerations, and data collections and analyses are just a subset of research skills acquired. In addition, all experimenters engaged in professional-quality presentation of their projects, including research objectives, research approach, findings, and contributions, also encouraging the ability to communicate, justify, and defend their research decisions and activities.

Although most of the experiments from this project are designed and conducted by doctoral students, ample opportunity for learning was available for both masters and undergraduate level students. System development activities were required for most of the research areas and many masters and undergraduate students were able to get hands-on experience in software development.

Goals, Objectives, and Targeted Activities

This project has three overriding goals:

1. Examine the issues of representation dimensions and structures for visualization including 1D (linear), 2D (hierarchy, network, or map), 3D (cube, VRML fly-through).

2. Examine viewing/zooming techniques: graphical fisheye view and fractal view.

3. Embed the analysis and visualization techniques in a dual-agent design that consists of a group agent and a personal agent.

Goals one and two have been completed with work currently focusing on refining the group agent and personal agent. Additional work includes a user evaluation of the agents using protocol/content analysis and interviews.

Project References

CI Spider and MetaSpider are available as downloads at: http://ai.bpa.arizona.edu.

Area Background

In the past decade, there have been many exciting research developments in organizational computing and information systems. As a new class of networked information resources has occurred, they have opened many new avenues of research for computer scientists, information scientists, and sociologists. On the other hand, in addition to drastically improved computing infrastructures, new computing techniques such as mathematical pattern analysis, neural networks computing, inductive machine learning, intelligent agents, and advanced information visualization have provided new light and insight into organizational computing. All of these emerging applications and techniques are closely intertwined with real-life problems, large-scale collections, and actual users in an organizational context. In sum, new applications and techniques have caused a dramatic shift in the computing paradigm from a tool-centric perspective (build it and they will come) to an organization-centric perspective.

Area References

H, Chen “Collaborative Systems: solving the vocabulary problem,” IEEE Computer 27(5):58-66, Special Issue on Computer-Supported Cooperative Work (CSCW), May 1994.

H. Chen, P. Hsu, R. Orwig, L. Hoopes, and J.F. Nunamaker “Automatic Concept Classification of Text From Electronic Meetings.” Communications of the ACM 37(10):56-73, October 1994.

C.A. Ellis, S.J. Gibbs, and G.L. Rein “Groupware: some issues and experiences.” Communications of the ACM 34(1):38-58, January 1991.

J.F. George, G.K. Easton, J.F. Nunamaker and G.B. Northcraft “A Study of Collaborative Group Work with and without Computer-Based Support.” Information Systems Research, 1(4):394-415, December 1990.

J.F. Nunamaker, R. Briggs and D. Mittleman. Electronic Meeting Systems: Ten years of Lessons Learned in Groupware: Technology and Applications. Prentice Hall: Englewood Cliffs, NJ, 1995.

Potential Related Projects

We would welcome the opportunity to collaborate with others who have an interest in this area.