Room 8 Schedule
Ford 202
Wednesday, April 15th, 2026
-
The Community Closet project is designed to address the problem of unnecessary individual expenses by creating a collection of community items to borrow and lend. This reduces cost for the community by participating in a reciprocal relationship of lending and borrowing. The project will serve as an intermediary management tool for keeping track of user data, finding items near the user, and connecting users for item sharing. The tool will also incorporate item condition reporting and user borrowing history to maintain a positive, safe experience. This project hopes to contribute to building community resilience and promoting a circular economy.
Faculty Sponsor: Fred Agbo
Discipline: Computer Science -
The Bishop Wellness Center that’s available to all current Willamette students regardless of their insurance coverage. However, the only way for a student to book an appointment is by talking directly to a staff member during the center’s hours of operation. There is no effective way to remind students of their appointments without the student manually putting it in their calendar.
For this personal project, I worked to build a web application that will allow students to book appointments online, and keep track of upcoming appointments. My hope is that this application will make our Wellness Center even more accessible.Faculty Sponsor: Fred Agbo
Discipline: Computer Science -
New England has some of the fastest rising temperatures in the world, and native tree populations are suffering because of it. This project investigates how temperature and other climatic variables impact tree presence across New England as well as tree growth and mortality in the Harvard Forest. We used a combination of predictive models (generalized linear model with zero-classifier, random forest, etc.) and causal analysis to measure tree resilience and habitat preference by species and location. These findings help inform future reforestation and conservation efforts amid a warming climate.
Faculty Sponsor: Zechariah Meunier
Discipline: Data Science -
Mountain pine beetles are a prevalent threat to western North American forest health. Due to how these beetles lay eggs in the bark of trees, disrupting nutrient flow and slowly killing trees, mountain pine beetle outbreaks can kill entire forests. Though these outbreaks have been studied considerably, action is difficult to take on a large scale. We used time series analysis, predictive modeling, and other data wrangling techniques to discover patterns in the beetle prevalence with the intent of informing future action against these beetles.
Faculty Sponsor: Zechariah Meunier
Discipline: Data Science -
This project explores temporal trends for streams in the watershed of Salem, Oregon, based on paired continuous monitoring sites operated by the City of Salem Public Works Department over nearly two decades (2006 – 2025). This project seeks to answer two main questions: (1) How have temporal trends in temperature and other water quality metrics evolved in Salem streams over the last two decades? (2) How do trends at upstream monitoring sites differ from trends at downstream monitoring sites? Our results have important implications for stream health and restoration practices in Salem.
Faculty Sponsor: Zechariah Meunier
Discipline: Data Science -
Agricultural and industrial runoff are a near-constant threat to ecosystem health and diversity. Chemical pollutants like sodium nitrate (NaNO3) and sodium sulfate (Na2SO4) are measured in freshwater ecosystems at increasingly elevated concentrations. Our project analyzes experimental data of the impacts of NaNO3 and Na2SO4 on aquatic organisms like mussels and frogs. The adverse effects from chemical exposure range from slower growth and metamorphosis to mortality. As with most environmental interactions, the relationships between pollutant concentrations and organism responses are nonlinear. Our results document the damaging effects of common chemical pollutants when they surpass the thresholds of important freshwater species.
Faculty Sponsor: Zechariah Meunier
Discipline: Data Science -
Wildfires in northern California have burned at high severity (the effects of a fire on the vegetation and soil) due to climate change and historical fire suppression. In response, state and federal agencies have implemented fuel reduction treatments, but treatment placement and timing are insufficient to reduce wildfire severity. This project aims to evaluate the effectiveness of existing fuel reduction treatments by quantifying where and when treatments reduce severity and where they fail, identifying gaps in wildfire mitigation strategies. We implement machine learning techniques to predict effective treatment locations to support fire agencies and indigenous tribes throughout northern California.
Faculty Sponsor: Zechariah Meunier
Discipline: Data Science -
Deforestation has become an increasingly severe problem over the last few decades, and mining is a difficult-to-control category of tree loss. In our analyses, we investigated tree loss at a global scale, and then focused on Madre De Dios, Peru, where deforestation due to mining operations has been a concern in once-protected forests. We performed various regression and additive models to relate tree loss to specific explanatory variables. Through statistical analysis and modelling of global and local deforestation and carbon data, we found several feedback loops that affect countless social and environmental systems. Our results suggest further work for environmental monitoring and enforcement of mining regulations.
Faculty Sponsor: Zechariah Meunier
Discipline: Data Science -
This project examines the expansion of the Sahara Desert and its impact on the Sahel in Senegal and Chad. Using NASA MODIS satellite data (2000–2025), we analyzed changes in land cover, vegetation health, and surface temperature to track desertification and environmental stress. Markov chain transition matrices were used to model the likelihood of land shifting between vegetated and barren states, while a random forest model predicts land cover based on vegetation and temperature patterns. By combining data science and environmental analysis, this study evaluates desertification trends and their implications for food security, economic stability, and restoration efforts in sub-Saharan Africa.
Faculty Sponsor: Zechariah Meunier
Discipline: Data Science -
Natural disasters are often fatal events that can cause millions of dollars' worth of damage individually. This Python Data Science project analyzes natural disaster data to identify trends, such as which countries are the hardest hit (in terms of both number of disasters and number of people affected) for both today and possibly the near future. Types of natural disasters analyzed include droughts, storms, floods, wildfires, landslides, and extreme temperatures. The results of this project will be revealed and discussed.
Faculty Sponsor: Haiyan Cheng
Discipline: Data Science -
This project quantifies and visualizes Oregon wildfires using data from the Oregon Department of Forestry and from presentation and research compiled by Dr. Bob Zybach. Both forest management practices and changes in state climate are related to trends in the quantity of wildfires and the total burned acreage in the state of Oregon. The purpose of this data science project in Python is to better quantify trends in Oregon wildfire data and to clearly visualize the results to present to a general audience. The results of data visualization and modeling will be discussed.
Faculty Sponsor: Haiyan Cheng
Discipline: Data Science -
The goal of this project is to train a Deep Neural Network to classify images into 10
groups based around broad types of plants and animals. It will take an already existing model and retrain it to perform this classification with a dataset of about 10,000 images. The ability of the resulting model will be revealed and discussed with examples of correct classification. We will also show confusion tables and calculate general accuracy rates.Faculty Sponsor: Haiyan Cheng
Discipline: Data Science -
Dating apps have fundamentally changed the way people form romantic connections. Despite the prevalence of these dating algorithms, what actually makes a profile successful is still not well documented. This project analyzes anonymized data from dating app profiles obtained through Swipestats to begin answering that question. By examining demographic and behavioral data such as swipe patterns, match rates, and messaging activity, we explore which traits and habits are most associated with match success. Visualizations and machine learning techniques built with Python guide our analysis and help pinpoint what predicts higher match rates.
Faculty Sponsor: Haiyan Cheng
Discipline: Data Science