​22nd Annual Undergraduate Research Symposium
​
School of Mathematics, Computing, Physics and Astronomy

​Analyzing the Postman Problem
​Analyzing the Postman Problem Oral Presentation | Discuss via Zoom
By: Emily Ziech
Faculty Sponsor: Andy Felt
The postman problem (also known as the Chinese postman problem) entails finding the shortest complete cycle around a graph which traverses each edge at least once. This problem is applied to postal routes as mail carriers must traverse each street so they deliver mail to every house on their route. This research uses actual postal routes from the Appleton post office. Both undirected and directed graphs are applied and investigated. Two models for undirected graphs are compared in their speed and effectiveness. A network flow model was used to analyze undirected graphs. AMPL is used to find the minimum number of extra times each edge must be traversed to make a complete cycle and Python is used to find the actual path along a directed graph, implementing Dijkstra’s algorithm.


Bulge-Disk-Bar Decomposition of Early-Type Spiral Galaxies in Galaxy Pairs
​Bulge-Disk-Bar Decomposition of Early-Type Spiral Galaxies in Galaxy Pairs Poster | Discuss via Zoom
By: Miranda Gorsuch
Faculty Sponsor: Adriana Durbala
We explore the relative role of “nature versus nurture” (intrinsically versus environmentally driven influences) in shaping the morphology and evolution of galaxies by performing a detailed photometric analysis of early-type spiral galaxies found in galaxy pairs using a Fortran code called BUDDA (Bulge/Disk Decomposition Analysis). We use red (i) filter images from SDSS (Sloan Digital Sky Survey). The main components of a spiral galaxy (bulge, disk, and bar) are modelled with appropriate mathematical functions. We present the scaling relations between parameters that describe each component of the galaxy. This analysis will allow us to gain more insight into the formation and evolution of spiral galaxies.

​​Exploring Properties of Galaxies and the Baryonic Tully-Fisher Relation
​​Exploring Properties of Galaxies and the Baryonic Tully-Fisher Relation Poster | Discuss via Zoom
By: Emily Ziech
Faculty Sponsor: Adriana Durbala
We investigate the properties of ~ 30,000 galaxies using data from the ALFALFA-SDSS Galaxy Catalog.  This is a radio optical catalog based on 100% complete Arecibo Legacy Fast Arecibo L-band Feed Array (ALFALFA) survey and Sloan Digital Sky Survey (SDSS). The baryonic Tully-Fisher relation, which describes the relationship between the mass of a spiral disk galaxy and its rotational speed, is examined. Other properties explored include stellar mass, star formation rate, absolute magnitude, and color.  These properties and their relationships to each other are analyzed and compared to those from previous studies found in research articles.

​​Exploring Quasars Through Their Broad Emission Line Shifts and Radio Morphology
​​Exploring Quasars Through Their Broad Emission Line Shifts and Radio Morphology Poster | Discuss via Zoom
By: Nicholas Quisler
Faculty Sponsor: Sebastian Zamfir
Quasars are the most energetic types of Active Galactic Nuclei, presumably powered by supermassive black holes accreting matter from their immediate vicinity. Their copious energy output originates in a relatively small volume of space, much smaller than the distance that separates the Sun from its nearest stellar neighbor. Given their cosmological distances and their compact physical size, the only hope to resolve their structure relies on spectroscopy, rather than direct imaging. We investigate large samples of optical quasar spectra, originally acquired by the Sloan Digital Sky Survey and subsequently measured and cataloged by various professional groups. We use very recent, vetted catalogs of spectral measures (publicly available), and extract their radio morphology maps from databases produced by the Very Large Array of Radio Telescopes at 20 cm wavelength. The focus is on the shifts (relative to the internal rest-frame of the quasars) of the characteristic broad emission lines (Balmer lines, MgII l2800 Å, etc.) and the extended radio-morphology driven by large-scale jets launched by the active galactic engines. We report on new insights into the fascinating world of quasars using a multi-wavelength approach.

​​Using Natural Language Processing to Support Curriculum Content Analysis and Visualization
​​Using Natural Language Processing to Support Curriculum Content Analysis and Visualization Poster | Discuss via Zoom
By: Hunter Frey
Faculty Sponsor: Tomi Heimonen
When assessing and revising the curriculum of an academic program, it is important to consider many factors. One element of interest is the overlap between courses in the program in terms of topics and learning outcomes. Overlaps can indicate both learning paths from introductory content to more challenging content or highlight redundancies. This project investigated the use of natural language processing (NLP) tools and techniques for partially automating the analysis of overlaps and connections between courses.

A proof-of-concept application was built to extract course descriptions and learning outcomes from course syllabi documents. Latent Dirichlet Allocation (LDA) was used to model course topics as sets of representative terms. On-going and future work includes the implementation of visualizations to represent relationships between courses based on the topic models. The results of this project could help academic programs in gaining insights into the content-based relationships in their curricula.

​ ​​ ​