ZOOM link for off-campus community members
The industry of sports gambling has exploded in the past years, and with it comes a new method to beat the odds: Data Science. Our goal is to create a program that allows anyone to set up a hypothetical MLB game and predict the winner. We'll be using a neural network and making a more rudimentary game simulation to predict the outcomes of hypothetical games. Final deliverable will be fully flushed out with a simple UI.
Faculty Sponsor: Haiyan Cheng
Discipline: Computer Science
PenFlow is a clean-looking, practical multi-platform note taking website designed to allow students to take notes in a freeform manner, allowing drawings, text, and the insertion of images. Our website is different from other note taking services because of how collaborative and freeform it is. PenFlow is designed for flexibility, perfect for an academic environment. Using Render to host our website, we are building it with Python, HTML, CSS, and SQL.
Faculty Sponsor: Haiyan Cheng
Discipline: Computer Science
As Willamette’s student population grows, the need for accessible event information has become more important. Current platforms such as Today@Willamette and outdated bulletin websites lack comprehensive updates and are often overlooked due to their formats. Physical mediums like Toilet Paper are limited in size, only sporadically updated, and fail to meet students' needs as many miss out on-campus events. Bearcat Board addresses this by providing a user-friendly interface for posting and discovering events, facilitating student engagement, and enhancing campus life.
Faculty Sponsor: Haiyan Cheng
Discipline: Computer Science
The purpose of this project is to understand and experience cutting-edge workflows and paradigms within the modern software development industry. By creating a game that simulates the software industry while engaging with the processes at every stage of development, the final product - a playable 2D game- will be informative and inspired by first-person experience. The results, lessons learned, and the future of the project will be discussed.
Faculty Sponsor: Haiyan Cheng
Discipline: Computer Science
Genetic algorithms can be used to solve optimization problems that do not have established solutions. Previous work has shown genetic algorithms to be effective at university course scheduling optimization. In this work, we have attempted the development of an app that can apply a genetic algorithm to the problem of course scheduling for the Chemistry Department at Willamette University. An individual’s calculated fitness was reduced by instances of double-booking, under/overscheduling, and insufficient class availability. The effectiveness of the app’s genetic algorithm was determined via comparison to the schedule produced by manual heuristics.
Faculty Sponsor: Haiyan Cheng
Discipline: Computer Science
We're tackling the problem of scarce mushroom data online, which makes it hard to create new software like mushroom recognition apps. We gather, organize, and improve data from different sources automatically, creating a big mushroom dataset. We then build an easy-to-use iOS app to show this data, following user-friendly design rules. Our project demonstrates how we can gather and share knowledge effectively.
Faculty Sponsor: Haiyan Cheng
Discipline: Computer Science
This project aims to predict Premier League player salaries using a comprehensive approach that incorporates machine learning and statistical modeling. Leveraging a vast dataset encompassing player performance metrics and player contracts the model employs advanced algorithms to analyze and identify key factors influencing salary variations. By considering variables such as goal-scoring prowess, assists, position, and player age, the project strives to unveil hidden patterns and correlations. The ultimate goal is to learn more about modeling and machine learning and apply it to a real-world application that aligns with our interests.
Faculty Sponsor: Hank Ibser
Discipline: Data Science
This project is an analysis of a data set of YouTube video statistics. This dataset was collected using the YouTube API, and includes the videos YouTube considers to be “trending”, based on their internal considerations. The data set includes information to identify the video, including the title of the video and channel, and data on the views, likes, and other quantitative information about the videos. We applied various models to this data to find significant relationships between the variables.
Faculty Sponsor: Hank Ibser
Discipline: Data Science
This project examines the analytics recorded by Spotify, especially those of loudness and genre, to determine the effect they have on the popularity of songs. This project utilizes techniques of simple linear regression, multiple linear regression, regression trees, and polynomial regression in order to come to a conclusion on how these factors affect song popularity. This analysis provides insight into why songs of different genres are popular and what attributes of certain genres cause those songs to be more popular than others.
Faculty Sponsor: Hank Ibser
Discipline: Data Science
In our NBA salary projection machine learning project, we wrangled our data, homing in on key performance indicators critical to salary determination. The model strategically incorporated a mix of numerical and categorical variables like points per game, rebounds, minutes played, and age category. Through the lens of linear regression, we harnessed the predictive power of these specific metrics, delving deep into their interplay to construct a coherent framework. This approach not only refines the accuracy of salary projections but also unravels the intricate dynamics underpinning the financial valuation of NBA players, offering a detailed and informed analysis.
Faculty Sponsor: Hank Ibser
Discipline: Data Science
A malignant tumor in the brain is a life-threatening condition. Glioblastoma is both the most common form of brain cancer and the one with the worst prognosis, with median survival being less than a year. The presence of a specific genetic sequence known as MGMT promoter methylation is a strong indicator of responsiveness to chemotherapy. Currently, the only way to identify this sequence depends on an intensive biopsy and weeks of analysis. This approach uses 3D convolutional neural networks and other methods to identify the cancer through imaging, hopefully improving cancer management and survival prospects for patients.
Faculty Sponsor: Haiyen Cheng
Discipline: Data Science
Through the use of RStudio, we will be exploring the significance of score and completed count. Along the way we will also explore the possibilities of other related variables.
Faculty Sponsor: Hank Ibser
Discipline: Data Science
This project is intended to use machine learning techniques to uncover trends in global biodiversity around the world and across different ecosystems and taxonomic groups. Using data from BioTIME that compiles the results from hundreds of biodiversity surveys, we tried to find patterns in the staggering complexity of the natural world and the systems we use to study it.
Faculty Sponsor: Hank Ibser
Discipline: Data Science