Adam Reagan
Department of Computer Science
Appalachian State University
Medical images contain a vast amount of information.
These images are often stored in digital format on computer disks
for further study and analysis.Some of this information is hidden
and not easily discernible. In the past decade, significant progress
has been made in knowledge discovery from a wide variety of large data
sets. Previous research in this area show that both wavelet lifting
schemes and neural network based data mining techniques can be used
to detect cancer in seemingly clear images. This talk discusses how
wavelet lifting schemes and 3-dimensional shape histogram sectoring
techniques can be used to classify images based on their content and
identify potentially cancerous regions in 2-dimensional mammogram
images.
Peter B. Henderson, Head
Department of Computer Science and Software Engineering
Butler University
This presentation will identify and motivate the topics to be included
in freshman discrete mathematics, discuss curricula issues, present
evidence that teaching discrete mathematics and problem-solving early
is beneficial, and discuss ways in which mathematical concepts can be
integrated and reinforced throughout undergraduate computer science
and software engineering curricula.
Stevan Kominac
Department of Mathematics and Computer Science
Dickinson College
The goal of this research is to evolve neural networks with a genetic
algorithm in order to investigate the circumstances, if any, under which
neural networks evolve as small-world networks. The fact that small-world
networks appear so frequently in nature, where they have emerged through
the process of evolution, is the motivation behind this research. Hence,
the question is, if we evolve neural networks through a genetic algorithm
that mimics natural evolution, could neural networks also evolve as
small-world networks?
Trevor Davis
Anne Maiale
Department of Mathematics and Computer Science
Dickinson College
Applying Something I Love to School (Trevor Davis):
During my internship with the Harrisburg Horizon semi-pro basketball team,
I was able to combine my love for basketball with my web development skills.
I designed and updated both the team website as well as the league website,
all while earning credit for class.
James Hamblin
Assistant Professor of Mathematics
Shippensburg University
One of the most common applications of the ideas of probability is to
games. However, in many games, the same action is repeated many times:
a die is rolled, a card is drawn, a coin is flipped, and so on. People
have certain expectations, knowing the probability, for how many times
we would need to repeat these actions to get a certain number of desired
outcomes. We will explore these ideas and how they apply to some common
(and some uncommon) games.
Beth M. Campbell Hetrick
PhD Candidate
Department of Mathematics
Bryn Mawr College
This talk provides a basic introduction to metric spaces. We
will define a metric space and consider examples of several different
metrics. We also consider unit balls. Finally, we mention some
applications of metric spaces.
Dr. Jeffery Forrester
Research Instructor
Vanderbilt University Medical Center
Innovations in genomics and proteomics have revolutionized the collection of
biological data, allowing researchers to generate huge data sets with a
reasonable expenditure of resources. Cluster Analysis provides an elementary
statistical tool for exploring these data sets, assisting in the
identification of internal data structures that would be difficult to find
without computational assistance. In this talk, we explore the concepts of
metrics, similarity/dissimilarity matrices, clustering routines, and
dendrogram construction. Examples are taken from biological signaling
networks and the mass spectral analysis of cancer biopsies.
Chats from previous semesters: