5/6/2016 Announcement of next year's Q-Club officers. Last Q-Club for the semester and
Hannah Durant will be presenting her SYE research
Title: "Graph Theory and Epidemiology: Using network models to represent and analyze the spread of infectious diseases"
Abstract: Throughout our history, humans have been plagued with devastating epidemics of infectious diseases. Recent events such as the Ebola epidemic in West Africa and the well known worldwide HIV/AIDS epidemic, have brought the threat of epidemics into current news and shed light on the vital field of epidemiology. How can we predict and prevent these outbreaks? If an epidemic does occur, how do we minimize the damage? Researchers have used network models, rooted in the discipline of graph theory, to find answers to these important epidemiological questions. This presentation will provide a brief introduction to network models and detail their advantages and limitations by analyzing two case studies on Ebola and SARS.
4/15/2016 Edward Morreale (Computer Science)
Title: "Videogames and you! A look at small scale videogame development"
Abstract: Videogames have been played and enjoyed for decades, but just how exactly is one constructed? In this presentation I will be giving my experiences in video game development. There are many different processes that are involved when even doing simple things like getting a character to move across the screen, and this can overwhelming to someone looking in from the outside. The goal of the presentation is to give a high level overview of the game development process, and looking at how some of the different components of games relate to the classes I have taken here at St. Lawrence. Additionally, I will talk about what are some practices that both professionals and other independent designers conduct. Other non-programming aspects of game development such as art and music will be discussed as well. The talk will conclude with a demo of my current progress.
3/11/2016 Robin Lock, Burry Professor of Statistics
Title: "CHODR – A Statistical Model to Rate College Hockey Teams and Predict Game Outcomes"
Abstract: Abstract: The College Hockey Offensive and Defensive Ratings (CHODR) is a system for ranking teams and predicting results of future games based on results of past games after accounting for the quality of opponents. Each team is given an offensive rating reflecting their scoring rate and a defensive rating to measure success at preventing opponents from scoring. Current CHODR results for men’s and women’s NCAA Division I teams can be found at it.stlawu.edu/~chodr. We discuss how these ratings are determined using a Poisson model for scoring and how they can be used to forecast future games –in particular giving a forecast for when St. Lawrence faces off against Clarkson in the ECAC playoffs this weekend.
2/26/2016 Bailey O’Keeffe
Title: "Oscars Study Examines Lack of Female Representation in Best Picture Nominated Films"
Abstract: Due to criticism about the gender discrepancy in Oscar-nominated films, we analyzed the screen time of male and female leads in films nominated for Best Picture from 2006-2014. This talk will focus on the differences found between lead screen times according to year and director gender, the relationship between how long a female lead is on screen and how likely it is that a film will win the Oscar, and also the lack of racial diversity in these films in light of this year's #OscarsSoWhite.
2/12/2016 Dr. Ed Harcourt
Title: " A Virtual Machine for Accelerating Database Joins Using a General Purpose GPU”
Abstract: We demonstrate a speedup for database joins using a general purpose graphics processing unit (GPGPU). The technique is novel in that it operates on an SQL virtual machine model developed using CUDA. The implementation compiles an SQL statement to instructions of the virtual machine that are then executed in parallel on the GPU. We use the three dimensional structure of the CUDA grid and thread model to perform a join on up to three relations at a time. Query execution results in speedups of 2 to 60 times on consumer-level GPUs depending on the size of the result set.
1/29/2015 Dr. Michael Schuckers
Title: “Fulbright and Finland”
Abstract: In this talk, I will introduce the Fulbright program including opportunities for students as well as talk about my experiences as a Fulbright Scholar in Finland. “The Fulbright Program is the flagship international educational exchange program sponsored by the U.S. government and is designed to increase mutual understanding between the people of the United States and the people of other countries.” (http://eca.state.gov/fulbright) Since 1946, the Fulbright program has awarded scholarships for students and faculty to study, teach and research in another country. In 2013, I was a Fulbright Scholar at the VTT Research Institute in Espoo, Finland working with their bioinformatics group. In addition to discussing my experiences with the Fulbright program, I’ll talk a small bit about the research I did at VTT on metabolonomics.
12/4/2015 Son Vuong
Title: "Haggling with Google Flight: Predicting Flight Prices"
Abstract: In today’s internet Age, travel agent is a profession of the past. So, how do you haggle with that flight search engine to get the best price on your flight home? This talk focuses on the relationship between the percentile of flights’ prices and various search options that a user can configure. The target variable, predicted percentile, shows where a particular query configuration lies in the range of historical prices. By using this variable, we would be able to determine whether that particular price is the deal of your life! Queried via Google Flights Engine on a fixed set of destinations, our data consists of over 2 million flights. We explored this complex relationship on three different levels: a fixed trip with given origin, destination, departure date and return date; a flexible trip given a range of departure date and return date; and a vacation trip given the fixed trip length of 9 days, that occurred in the next 300 days
11/13/2015 Sydney Bell and John Tank
Title: "Sports Analytics Internships"
Abstract: We’ll be taking a look at the sport statistic industry as it relates to both the team aspect as well as the business side. Sydney Bell worked with the Florida Panthers, and will be discussing some of the analytic problems she worked on, how she got the internship, and what she learned while commutating with staff members within the organization. John Tank worked on the business side, as he was a data analytics intern for a company called CoachMePlus. During his time there he worked mainly with data collection, analysis of client performance, and identifying trends within data. They will also say what they learned from their experiences, and give advice to students interested in the field.
10/23/2015 Professor Choong-Soo Lee will be presenting
Title: "Rise of the Bots: Bot Prevalence and Its Impact on Match Outcomes in League of Legends"
Abstract: Do you play League of Legends or know family members and friends who do? Have you ever experienced or heard of accusing other players of being a bot? Guess what! They do exist and have an impact on match results! Come and find out how we estimated the prevalence of bots and determined their negative effect on League of Legends matches.
10/2/2015 Evan Smith will be presenting
Title: "Reducing L-Band Wide Observations of Optically Selected Galaxies"
Abstract: Observations of galaxies in the Virgo Cluster were completed at the Arecibo Observatory in the spring and summer of 2015. 161 targets were observed, selected by criteria such as magnitude and shape from the Sloan Digital Sky Survey. The targets, which were too dim to be detected by Arecibo’s ALFA drift scanner, were observed with the L-Band Wide detector. Once reductions in an IDL environment were done, these data were matched to the targets from the Sloan Digital Sky Survey and the GALEX/MAST catalog. Comparing the galaxies that were detected against the galaxies that were not detected (by the L-Band Wide receiver) will allow us to refine our method of choosing HI-rich galaxies in the 2000km/s to 9000km/s range and prepare for the Arecibo Pisces-Perseus Supercluster Survey, which will use the same method of target selection. 115 of the 161 targets observed had positive detections, a 71% success rate.
9/18/2015 Janelle Frederics (Statistics Major - Class of 2016
Title: Exploring Carbon Density Loss in the Pantropical Forests
Abstract: During my summer fellowship at St., Lawrence University, I have been working on creating an interactive web app titled Exploring Carbon Density Loss in the Pantropic Forests. There has continuously been an increase in human activities, such as industry, transportation and electricity, which has caused instability in our eco-system and reduced the productivity of forests. Work on this topic uses the term edge effect to measure this reduction in productivity. I have been provided with information on the edge effects present in pantropic forests by the Natural Capital Project, a research group housed at Stanford University, whose general mission is to help preserve the environment we live in. Through the computer program R, and its subsequent packages, I created a web app for this group so scientists who use this data have an easy way to visualize and use the information that is pertinent to them. Come see the web app I created and see the opportunities that are available to all students during the summer fellowship program. A preview of this app can be found at: http://shiny.stlawu.edu:3838/NatCap/CarbonLoss
9/4/15 Math, Computer Science and Statistics Faculty will be presenting.
Title: Where’s our pizza?! (Welcome back!)
Abstract: Welcome back (or just welcome) to another year of Q-Club! We will kick this year off with a general info meeting about the department – including meeting the Q-Club officers, the new Math professor, and hearing about some of the fun and exciting events the Mathematics, Computer Science, and Statistics Department has to offer this semester! And don’t worry, as always, pizza and snacks will be provided.
4/17/2015 Jenna Street
Adjusting Grade Point Average for Course Difficulty
Abstract: Class rank is a measure of a student's performance compared to the performance of other students in his or her class. The conventional method for determining a student's grade point average involves the earned grades and the number of units or credits that the student receives from each course. However, this process does not include a way to incorporate course difficulty, which can vary due to the nature of the course material or the particular grading standards of a professor, among other defining characteristics. The purpose of this project is to analyze ten years' worth of anonymized grade history for St. Lawrence University and build a model that provides a student's grade point average after adjusting for course difficulty. By doing so, we are able to re-rank students within each class in order to more accurately provide a comparison of classmates. We find that incorporating course difficulty into the ranking process impacts the order of class rankings though the biggest factor impacting class rankings is the individual student.
4/3/2015 Colleen Bradley
"Could Your Sleep Schedule be Affecting Your Mental Health? An Application of Mixed Effects Modeling"
Abstract: Mixed effects modeling refers to a broad array of techniques that can be used to model data that violate the usual linear model assumptions. We will use mixed effects modeling techniques to analyze data from a sleep study previously conducted by the psychology department. The study tested the effects of changing the beginning start time of the school day for high school students to a later time through the use of self-report surveys. The current available research suggests that sleeping habits can have an influence on the mental health of an individual. It has been shown that insufficient sleep is often linked with poor mental health, especially in regards to depression, anxiety, and stress. We will use the data from the sleep study to discover if the change in start time of the school day will have an effect on the mental health of these high school students.
3/27/15 Nathaniel Shenton
"The Double Pendulum, A Case Study of Chaotic Behavior"
Abstract: The double pendulum is a dynamic system in which a second pendulum is connected to the first, allowing the second to swing freely. The motion of the resulting dynamical system will associated to the motion of the pendulum. This system experiences both periodic and chaotic tendencies. We will use this example to show what it means for a system to be considered chaotic. To do this we will explore the known visual tools that can help detect chaos, including an actual built pendulum.
3/6/15 Kirby Kaylor
"Title: Old McDonald had a Sensor"
Abstract: High tunnels are gaining popularity in farming communities as they allow for vegetables to be grown all year. When the outside temperature is too cold, the sides can be rolled down to capture the sun’s natural heat, while minimizing heat loss. When the outside temperature is warm enough, the sides can be rolled up to prevent overheating. These countermeasures require real-time temperature data from the tunnels. Sensor motes are battery-operated devices that can read data such as temperature and humidity and also form a wireless ad hoc network to transmit sensor data. A computer can collect up-to-date sensor data from the motes and control relevant equipment to regulate the temperature in the tunnels.
2/27/15 Dr. Sam Vandervelde
"Graphing Groups in the Projective Plane"
Abstract: In this talk we will introduce a novel method for visually presenting the group law for the integers mod m, i.e. the cyclic group of order m. Along the way we will meet the finite projective plane, arguably one of the most elegant geometric objects in existence, in which every pair of lines intersects precisely once. We will also discover that 14 is more unlucky than 13.
2/6/15 Brandon Lustig - will be presenting some work that he did with Dr. Robin Lock last fall on the use of statistics to rank PGA golfers.
"Head-to-Head Comparison Models to Rank PGA Tour Players"
Abstract: The current rankings systems for professional golfers, such as the official world golf rankings and the Fed Ex cup standings, put emphasis on winning tournaments and finishing near the top instead of looking at every round played with equal importance. To provide alternatives that are not so heavily weighted on winning tournaments, we use models, such as Bradley-Terry, that rely on head-to-head comparisons for all pairs of players for every tournament round. We discuss the details of finding ratings using these models, estimating the probability of a player beating another in a round, and compare the results using data from the 2013-2014 PGA tour season.
1/23/15 Boris Jukic and Joe Skufka from Clarkson University
"What's the big deal about big data?"
Abstract: The term `big data' is most commonly understood to mean massive volumes of diverse and rapidly growing data that are not formally modeled (mostly unstructured or lightly structured), from various sources such as smart devices, social media, or sensors, in a variety of formats such as blogs, emails, tweets, or any unstructured content in digital format such as text, video, or audio. There is a lot of hype and confusion regarding the true meaning and potential of this field and its uniquely inter-disciplinary nature. This talk will bring clarity to and debunk myths surrounding terms such as Big Data, Data Analytics and Data Science. (Such as "A data scientist is a statistician who lives in San Francisco.") The talk should be intriguing to a general audience and will hopefully inspire students to learn more about career prospects in this expanding field.
12/9/2014 Abby Ross and Dave DiStefano
"The most famous numbers you've never heard of"
Abstract: In this talk, we will introduce the Catalan numbers, their formula, and discuss an example to illustrate their recurrence. A central idea of this SYE was to understand what happens when we take this set of numbers and “bump it up” to a higher dimension. We wrote code, conjectured a formula, and then proved this formula for "tri-Catalan numbers” - Catalan numbers in three dimensions. We then generalized this formula to all dimensions. We will close by discussing additional research we have done in relation to the Catalan numbers.
10/21/2014 Mitchell Joseph on the Budapest Semester in Mathematics
"The Budapest Semester in Mathematics"
Abstract: Budapest Semesters in Mathematics is an academic program held in Budapest, Hungary for American and Canadian undergraduates. Initiated by Paul Erdős, László Lovasz, and Vera T. Sós, the program provides a unique opportunity for North American undergraduates. Through this program, mathematics and computer science majors in their junior/senior years may spend fall, spring or summer semesters in Budapest and study under the tutelage of eminent Hungarian scholar-teachers.
10/7/2014 Skyler Ng
"PHOENIX, a Helping Hand"
Abstract: With today's technological advances, people spend increasing amounts of time working on computers. To help people work efficiently, this project aims to create an intelligent software assistant that will help automate tedious commands that users often do on a desktop computer. The main task of this program is to search files or documents by the name or content. The software process can be broken down into three different sections. In the first section, the program will analyze an English command, tag each word with its part of speech, and parse it to determine its grammatical structure. The second section will focus on processing the parsed command into a command that the computer can understand. The final section will return the output to the user interface in a readable format. This program will be written in Python because it is a useful tool for string manipulation, has easy-to-read syntax, and contains the Natural Language Processing Toolkit (NLTK), which provides functions that deal with text processing.
9/9/2014 Dan Look, Associate Professor of Mathematics
"Can authorship of a contested piece be determined by statistics?"
Abstract: No, it can't. However, statistics can provide supporting evidence. Stylometry is the study and quantification of writing style, often using statistical methods. These techniques can be used in conjunction with more conventional analyses to suggest authorship of a contested work. Most notably, this was used to solidify the authorship attribution for the anonymous Federalist Papers and the 15th book of Oz. We'll discuss some past examples and some of the techniques involved.
9/23/2014 Danny Driscoll, Jenna Street and Son Vuong
"Careers in Actuarial Science"
Abstract: Actuaries are leading business professionals who manage the impact of risk and uncertainty. They use mathematical and analytical skills on the job at a variety of industries to help businesses protect themselves against loss. Come learn about the process of becoming an actuary, including the exam process and the classes SLU offers to satisfy requirements. Hear about students’ experiences at actuary internships this past summer and discover this exciting but relatively unknown career path.
This is the last Q-Club meeting for the semester.
4/29/2014 Johnn Balderston
"Foxes, Hipsters, and The Internet Meme: A First-World Social Epidemic"
Abstract: The Internet Meme: a fast spreading, sometimes “viral,” internet fad that is quite possibly the fastest mutating disease known to mankind. The Meme virus threatens the health and abdominal circumference of individuals everywhere. The multiple strains of the virus and its speedy mutation rate have left the grand majority of the human race perpetually infected. For this reason, we create a mathematical SIR model to demonstrate the spread of memes, where individuals can either be Susceptible (S), Addicted (A), or Rehabilitated (R). Our SAR model incorporates social impacts on the spread of this dreaded plague, including personal preference, hipster effects, boredom, and meme mutation. Observing the internet meme in this manner allows for the relevant understanding of social diseases, in which interactions within the population can result in a form of vaccination or devaccination, unseen in typical SIR modeling. Our hope is that our SARs will lend insight into combating the spread of this debilitating disease.
4/22/2014 Dan Mulcahey will provide an introduction to cluster analysis and discuss an application in Economics
"Identifying Groups of Similarly Performing Mutual Funds Using Cluster Analysis"
Abstract: Cluster analysis is an explanatory tool used to identify groups within data. After giving an overview of some clustering algorithms, we use recent percent returns to find clusters of similarly performing mutual funds and demonstrate how Word Clouds can be used to conveniently summarize fund descriptions and quickly identify the similarities among funds in each cluster, and the differences among funds between clusters.
4/8/2014 Brian Kane and Anton Stoychev will be talking about driving a remote control car with Raspberry Pi!
"Using Your Smartphone to Control a RC Car"
Abstract: Computers have seen innumerable advancements over the years, the most noticeable of which being a dramatic increase in power, somewhat paradoxically paired with an equally dramatic reduction in size. The Raspberry Pi mini-computer allows us to hold the power of a desktop quite literally in the palm of our hands. The power, size, and affordability of the Raspberry Pi make it ideal for use in small electronics projects with a “DIY” flavor. Using a bit of engineering, we were able to take advantage of the Pi’s power and size to directly attach it to a RC car. This allowed us to have easy access to the car’s motor and steering functions. We then used Python and the Android Software Development Kit (SDK) to implement a client/server architecture through which we remotely controlled the car using an Android smartphone.
3/25/2014 Eric Budge and Steve Petramale
"On Counting Triangles"
Abstract: Algorithms for counting triangles in massive graphs have been studied intensely. This fundamental tool has been utilized in network analysis for the computation of metrics including clustering coefficients and transitivity ratios. The degree sequence S of a graph G is a list of the degrees of the vertices in G in non-increasing order. A particular degree sequence S is called graphic if and only if a graph G can be created using S. This work focuses on graphs with degree sequences containing a unique triple, a sequence which has three repeated terms and all other terms distinct. Polynomial expressions are given which count the number of triangles in realizations of such graphic unique triple sequences.
3/4/2014 Spencer Timerman
"Teaching AI to Work Together"
Abstract: : Teamwork is hard. It's difficult for humans and its difficult for computers. In this talk we will address computer learning, particularly in the case of a team of learners. Our problem was a simple one: five locations, ten players, with the object of the game to have more of your players at a majority of the locations. We found that certain reward mechanisms worked more effectively than others to teach the learners , and that our agents could not learn against opponents that acted with a certain degree of randomness.
2/18/2014 Jack Holby
"Variations on the Euclidean Steiner Tree Problem"
Abstract: Imagine you are trying to connect a set of cities to a highway. What is the best way to do this? Further, imagine a set of cities wish to build a highway between them. Where is the best place to build it? The Euclidean Steiner Tree Problem (ESTP) attempts to create a minimal spanning network of a set of points by allowing the introduction of new points, called Steiner points. We'll discuss a variation on this classic problem by introducing a single “Steiner line” in addition to the Steiner points, whose weight is not counted in the resulting network. For small sets, we have arrived at a complete geometric solution. We'll also discuss heuristic algorithms for solving this variation on larger sets. We believe that in general, this problem is NP-hard.
2/4/2014 Sam Vandervelde, Associate Professor of Mathematics
The World's Hardest Elementary Domino Tiling Problem
Abstract: : Tiling problems have long delighted recreational mathematicians due to their potential for elementary formulation yet unexpected difficulty and elegance. Join Dr. V for a dizzying survey (literally---have you ever seen a Prezi presentation?) of classic and less well-known tiling puzzles. We'll try out some of these puzzles together, then conclude by examining more closely one such problem that appeared on the 2009 USA Math Olympiad, successfully solved by only a single student in the entire country.