Mary Anne L. Egan

Comparative study of a genetic fuzzy c-means algorithm and a validity guided fuzzy c-means algorithm for locating clusters in noisy data

The partitioning of data into clusters is an important problem with many applications. Typically, one locates partitions using an iterative fuzzy c-means algorithm of one form or another. Unfortunately, the results of these techniques depend on the cluster center initialization because their search is based on hill climbing methods. Recently, there has been much investigation into the use of genetic algorithms to partition data into fuzzy clusters. Genetic algorithms are less sensitive to initial conditions due to the stochastic nature of their search. In this paper we compare the two techniques when locating fuzzy clusters embedded in noisy data and discuss the advantages and disadvantages of both methods.

Locating clusters in noisy data: a genetic fuzzy c-means clustering algorithm

The paper investigates the use of a genetic algorithm to locate fuzzy clusters embedded in noisy data. The partitioning of data into clusters is an important problem with many applications. Typically, one locates partitions using an iterative fuzzy c-means algorithm. To overcome some of the shortcomings of fuzzy c-means, a genetic c-means clustering algorithm is implemented and evaluated. It was discovered that this genetic c-means algorithm performs well in the absence of noise. When the clusters are embedded in noise, the genetic algorithm is not as robust as the validity guided robust fuzzy clustering algorithm. The paper concludes with a discussion of what factors contribute to the performance and what modifications may increase the robustness of the genetic c-means algorithm.

Students with Asperger's syndrome in the CS classroom

More students with Aspergers Syndrome are entering higher education and a majority of these students are choosing a computer science curriculum. With an increasing number of students with disabilities in the classroom, it is important to understand the particular aspects of a disability that will affect classroom situations. This paper presents common characteristics of students with Aspergers Syndrome and offers techniques that will maximize the students educational experience and minimize disturbance to other students.

Outreach programs to promote computer science and ict to high school and middle school students

Catherine Lang Swinburne University of Technology Information and Communication Technologies Australia +61 3 9214 5884 clang@swin.edu.au Reyyan Ayfer Bilkent University Ankara, Turkey +90 312 290 5065 ayfer@bilkent.edu.tr Annemieke Craig Deakin University School of Information Systems Deakin University, Australia +61 3 5227 2152 acraig@deakin.edu.au Jane Chu Prey Microsoft Corporation Microsoft Research Redmond, WA +11392145884 jprey@microsoft.com

Accessibility and computer science education

Brian J. Rosmaita (Moderator) Department of Computer Science Hamilton College 198 College Hill Road Clinton, NY, USA 13323 brosmait@hamilton.edu Katherine Deibel Dept. of Computer Science and Engineering University of Washington Box 352350 Seattle, WA, USA 98195 deibel@cs.washington.edu Robert F. Cohen Department of Computer Science University of Massachusetts Boston 100 Morrissey Boulevard Boston, MA, USA 02125 rfc@cs.umb.edu Mary Anne L. Egan Department of Computer Science Siena College 515 Loudon Road Loudonville, NY, USA 12211 maegan@siena.edu

A Java simulation tool for fuzzy clustering

Emerging technologies on the World Wide Web promise to make program, algorithm and concept simulations universally accessible, and Java appears to be the best technology available. Simulations involving animation and visualization have a tremendous benefit when applied to various algorithms. We present a simulation tool for experimenting with concepts in fuzzy clustering that has proved useful in visualizing the results and demonstrating the computation method of the algorithms. This is an asset when working with people unfamiliar with the mechanics of fuzzy clustering, such as non-computer scientists or students. This system was integral in the development of an algorithm capable of locating an unknown number of clusters embedded in background noise.

The Importance of Outreach Programs to Unblock the Pipeline and Broaden Diversity in ICT Education

There is a need for outreach programs to attract a diverse range of students to the computing discipline. The lack of qualified computing graduates to fill the growing number of computing vacancies is of concern to government and industry and there are few female students entering the computing pipeline at high school level. This paper presents three outreach programs that have the underlying assumption that students need to be reminded about the creativity and potential of computing so that it remains on the radar of their future career options. Each program instigated social and cultural change through a paradigm shift where girls moved from being ICT consumers to ICT creators. By exposing students to a wide variety of ICT activities and careers during secondary schooling, they were more likely to consider studying information systems, computer science or any other computing course at the university level. Results are presented showing student attitudinal changes as well as observed increases in enrolments at secondary school and university courses.

Perspectives on history in computing and education

Can history inspire and motivate students in computing programs? Can it help attract or retain students? History has always been part of academic learning. There are many examples of history-oriented courses in areas such as music, art, and mathematics. However, courses that focus on the history of computing are fairly rare and computing curricula recommendations of the past have often lacked this essential component of computing education. Indeed, computing courses often neglect the role of historical issues in their content. This is unfortunate, because there is much to learn from computing history. It broadens one’s perspective on the field and helps students explore the inner thinking of key historical figures and the events these individuals produced. History also enables students to learn from past events and discuss “lessons learned” in their program of study. In this panel, the presenters will argue for several approaches to using history as an interesting and valuable addition to the teaching repertoire for computing educators at both the precollege and college levels. The panelists will explore the importance of history in the computing curriculum and how to apply this theme in computing programs as a tool for learning. The panelists encourage computing educators to discover ways in which history can make the study of computing topics exciting. We hope, thanks in part to this panel presentation, that educators will gain insights regarding tools they can call upon to help them include the study of history in their computing courses.

Teaching a "women in computer science" course

There has been much research on the lack of women in the discipline of computer science. It is an important area about which we need to educate our students. Unfortunately, information about suggested syllabi, activities and textbooks is lacking. This paper presents the outline of a course that has been taught including the books, papers and activities that were used. It also presents the outcomes of the course and suggestions for future incarnations of this course.

FCLUST: a visualization tool for fuzzy clustering

Emerging technologies on the World Wide Web promise to make program, algorithm and concept simulations universally accessible. Simulations involving animation and visualization have a tremendous benefit when applied to various algorithms. We present a simulation tool for experimenting with concepts in fuzzy clustering that has proved useful in visualizing the results and demonstrating the computation method of the algorithms. This is especially advantageous in a classroom or laboratory setting where students may become more comfortable with the mechanics of fuzzy clustering through personal discovery and online experimentation.