Proposal -- An Analysis project of InfoVis 2004 Contest

1. Student names and email addresses:

• Ron (Jung-Rung) Han, MS Student, Computer Science Department jhan@cs.ubc.ca
• Chia-Ning Chiang, PhD Student, School of Library, Archival and Information Studies Chia_ning_chiang@hotmail.com

2. The domain, task, and dataset set by InfoVis 2004 Contest

• Domain: InfoVis 2004 Context ¡V The History of InfoVis, which is information visualization in the context of bibliometrics defined as the following:

Visualizations that support the discovery and identification of major research topics, relationships between members of the community, trends over time etc.

• Task: The goal is to solve these tasks with visualizations.

1- Create a static (non-interactive) overview of the 10 years of the InfoVis conference.

2- Characterize the research areas (areas/topics to be defined by you) and their evolution (your interpretation; time could be one choice).

3- Where does a particular author/researcher fit within the research areas?

4- What, if any, are the relationships between two or more or all researchers?

Additional related items to build into the visualizations include uncertainty, reliability, range, flexibility, broader applicability. We will try as much as we can do within the time frame to test the additional requirements. (see http://www.cs.umd.edu/hcil/iv04contest/info.html)

• Dataset: The dataset contains complete metadata for all the papers of 8 years (1995-2002) of InfoVis Conference and their references. The metadata includes publication title, authors, keywords, abstract, references, and links to original papers available in the ACM Digital Library. (see http://www.cs.umd.edu/hcil/iv04contest/register.html)

3. Personal expertise.

• Ron¡¦s expertise: graduated from UBC CS, Ron has experience with C, C++, Java, and OpenGL.
• Chia-Ning has some knowledge about bibliometrics, which is a type of research method utilizes quantitative analysis and statistics to describe patterns of publication within a given field or body of literature. Many bibliometric researches nowadays have been using visualization to produce maps for analysis presentation. InfoVis 2004 Contest¡XThe History of InfoVis encourages visualization designs that support the discovery and identification of major research topics, relationships between members of the community, trends over time etc., which provides a very good chance of both practicing and learning opportunities.

4. Proposed solution:

• This is a "survey" type project, where we try to apply different existing visualization packages (a MDS-program, such as VxInsight or SPSS/SAS MDS-module) on this dataset and discuss the pros and cons of each. If then we have a new way to visualize the dataset that cannot be found in existing programs, we can consider writing our own program to do it.
• The proposed infovis solution. http://www.slais.ubc.ca/chiang/cpsc533c/assignment/termproject.ppt
• The proposed implementation approach.

We will use existing softwares¡XVxInsight

.

The data import program ¡§VxImport¡¨ is not working for me. I will probably use Java to write a program to convert the XML into a tab-delimited file.

If we decide to implement our own algorithm, I will use Java.

• Milestones.

(1) Making the XML dataset into a Microsoft Access or Microsoft EXCEL data file containing tables according to the required tasks, such as:

¡³ Tables containing selected field

¡³ Tables containing fields sorted by frequency

¡³ Tables containing co-occurrence frequency and cluster pairs

¡³ Tables containing cluster results

¡³ Tables containing citing and cited pairs

(2) Preparing the VxInsight files

¡³ Construction of .DB, .CONFIG, .STOP files.

¡³ Construction of a REF file

¡³ Construction of a VxInsight File (.SIM)

¡³ Creating a coordinate file.

¡³ Creating a statistics file.

¡³ Creating a local cache file(.CACHE)

(3) Preparing a matrix file for VxInsight (or SPSS/SAS MDS)

(4) Tests and evaluations

(5) Add programming, if necessary

(6) Write an analysis report