CPSC-01 Turing Tested NPL Quora Avatar
Team Members: Adrian Abeyta, Anthony Tobias, Matthew Moore, Caleb Barker
Faculty Advisor: Scott Broder
Passing the Turing test for computer intelligence means a human is unable to determine if the question-and-answer results are generated by a machine or human.
Project: Create an Avatar using BERT or GTP-2 that generates answers to a limited category of questions on Quora questions that pass the Turing test as measured by the number of upvotes or comments by humans reading the posts generated by the Avatar.
CPSC-02 Machine Learning for Real Estate & City Planning, Phase III
Team Members: Ryan McKenzie, Zachary Sahlin, Ariana Jansma, Dominic Bevilacqua
Faculty Advisor: Michael Tobias
Sponsor: Pangeon LLC
There is a significant demand from public and private customers for the ability to predict a number of real estate related metrics. High profile failures in this field (e.g. Zillow, Opendoor) have not yet addressed the computational irreducibility of these complex systems.
Project:The team will utilize a variety publicly available data sources (e.g. population, prices, traffic) at multiple spatial and temporal scales and create a machine learning software application to predict actionable information that could be used by city planners or real estate professionals.
CPSC-03 Restructuring of Language Translation
Team Members: Jason Lunder, Riley Sikes, Leon Garcia-Camargo, Jaymin West
Faculty Advisor: Gina Sprint
Sponsor: Pangeon LLC
State of the art translation software (e.g. Google Translate) depends on Transformers (Deep multi-headed attention networks). This architecture was designed with the hardware in mind first and human language a distant second. GPT3, another popular language model, was trained on 1.1 million years of English but still makes mistakes that would seem silly to a 3-year-old. As a result, they appear to be the computational equivalent of a long-lived parrot.
Project: The team will attempt to complete a Neuro-symbolic end-to-end translator for English/Spanish that uses dependency trees for interlingua. Using off-the-shelf parsers and word embeddings, the team will build a visual labeling tool and use that tool to create the data needed to train Pangeon’s model.
CPSC-04 Groove Web Extension
Team Members: Nelly Alger, Katherine Stevens, Linden Beemer, Ethan Bao
Faculty Advisor: Daniel Lenz
Sponsor: Student Proposed
The overall goal of this project is to create a functioning Google Chrome Web Extension that rates the environmental sustainability of different online retailers. Groove’s goal is to give fashion retailers a rating of 1 to 10 in CO2 emissions, waste impact, and fair trade standards in order to make consumers more aware of the impact of their favorite clothing websites. These ratings will be determined using publicly available data from other watchdog websites and displayed to the user in a Google Chrome extension as well as on the official Groove website.
CPSC-05 ComSem App Development
Team Members: Zachary Ambroseo, Avery Edmondson, Dominic MacIsaac, Jared Torii
Faculty Advisor: Jasmine Jans
Sponsor: Dr. James Hunter
This project continues previous Senior Design projects (2015-2022) that created comsem.net, now a fully-functioning web-based application for English language learners to receive feedback on their oral language production, built with the Django database framework and MySQL back end, hosted on Heroku/AWS. The 2021-2022 team added improved security measures, error logging, and explored new machine learning neural networks (BERT) to assist in the automatic detection of grammatical and morphological language errors.
Project: The next phase involves the continued development of a range of discrete modules to enhance the existing platform including improved machine learning, automated self-assessment tools, statistical analysis, and user interface improvements. The Senior Design team can choose one or more of these, depending on their interests and experience, but all will involve developing their programming skills in multiple ways.
CPSC-06 HIPPA Compliance Gamification, Phase II
Team Members: Jakob Kubicki, Luke Martin, Kyle Manning, Ahmadullah Moltafet
Faculty Advisor: Mike Mudge
Medcurity’s goal is to help businesses train employees and manage the complex requirements of medical confidentiality and security.Last year, the Gonzaga team created an application to turn HIPAA compliance into game with the ability to have Administrator, Employer and Employee based roles. Users are assigned different categories and there are quizzes within each category. At the end there is a leader board that creates some friendly competition.
Project:Take this game to the next level with the following enhancements:
- Further “gamifying” the app (avatars, leader board enhancements)
- Enabling users to reset their passwords without intervention
- Enabling some “micro-learning” techniques
- Keep learner history instead of wiping it out
- Add more movement in the app (this may tie into gamification)
- Add depth and 3D to the background
- Have a way to have two (or more) different trainings that could be taken with separate questions like one that is solely HIPAA focused and one that is more cybersecurity focused, for example. (So users could be assigned to one or more trainings and then under those umbrella training assign the modules.)
CPSC-07 Air Quality Monitoring Data Harvest & Community Display
Team Members: Emily Inkrott, Daniel Balboni, Andrew Perez, Sarah Hagen
Faculty Advisor: Chris Sharman
Sponsor: Spokane Public Library
Spokane Public Library has purchased and will install an interior and exterior Purple Air II air quality monitoring system at each of its 7 new branches: Hillyard, Shadle Park, the Hive, South Hill, Indian Trail, Liberty Park, and Central (downtown).
Project:The team will create a database that integrates available historical data and integrates real-time data from new sensors. The team will then create a new portal to display this “Community Data” which will allow the data to be accessed and analyzed by the public and displayed in a custom interface at the Spokane Public Library entrances.
CPSC-08 Passenger Service App
Team Members: Alan Poblette, Samantha Blevens, Vy Nguyen, Brennan Longstreth
Faculty Advisor: Chris Sharman
Develop a new passenger service app intended to support new communications between the passenger and the airplane. Possible features could include:
- Food ordering
- Digital lavatory queuing
- Seat row market proximity sensor
- New UX for attendant call / reading light function
- Passenger air outlet control
CPSC-09 Research Project Database/Frontend
Team Members: Lin Ai Tan, Everett Johnson, Justin Salsbery, Emily Bodenbender
Faculty Advisor: Pete Messina
Project: Design an information system for storing, linking, and mining research and development artifacts from Boeing product development. System will integrate with tools from Boeing’s cloud infrastructure to provide a robust and modular architecture. System designers will consider what makes for an intuitive interface and how to deploy a system that can be maintained through a CI/CD pipeline.
CPSC-10 BioPath Graphical Learning Tools for BioChemistry
Team Members: Cole Stainsby, Zachary Burnaby, Benjamin Higley, Joshua Schmitz
Faculty Advisor: Bethany Alcamo
Sponsor: Dr. Jeff Watson
The goal of the project is to continue to develop an existing interactive, graphically-rich web app to help students understand the structure, organization and regulation of metabolic pathways in biochemistry.
Biochemistry is a dynamic science. Much of the content we discuss in a biochemistry course depends significantly on time-dependent concepts such as dynamic motion, structural change, flow of molecules through a series of coupled catalysts, and more. However, textbooks and even online tools rarely demonstrate this well, and when they do, they are not interactive; a student can’t change a variable in the system to see how the system responds.
The past three years, senior teams have developed a web app the has implemented the basics of the biopath tool, but this year will attempt an overhaul to add new features needed to include enriching the graphical complexity of the app, adding more detailed mathematical modeling of flow through pathways, adding ability to connect multiple pathways to one another, adding ability to handle cycles in a pathway and adding ability to have multiple substrates interact with enzymes and yield multiple products.
CPSC-11 Gonzaga Chatbot using RASA Framework
Team Members: Maya Fleming, Adam Kowalchyk, Hunter Banks, Joshua Vahlberg
Faculty Advisor: Cynthia Freeman
Sponsor: Cynthia Freeman
Project: Design and implement a working prototype of a general help chatbot for Gonzaga students. This prototype domain-specific Intelligent Virtual Assistant (IVA) should be able to source available data and answer any questions relevant to Gonzaga University such as:
- The location of certain facilities and perhaps provide maps or directions given a starting point
- The hours of certain eateries on campus like in the Crosby building
- COVID questions
- Graduation requirements
- Clubs to join
- Available Majors
CPSC-12 Aggregation of Disparate Medicare & Medicaid Data
Team Members: Cameron Williamson, Caden Kim, Hunter Hauser, Jacqueline Ramsey
Faculty Advisor: Gina Sprint
Sponsor: Amend Health
A tool has been created by Amend Health that at appropriate intervals, collects and aggregates reimbursement data from Federal Medicare data systems, and State Medicaid data systems, so that reimbursement data for given procedures can be analyzed and compared. This solution is cloud based, includes automated data collection tools, algorithms to compare disparate data, efficient data modelling, and a management interface to allow for control of the tool.
Project:The team will implement software engineering, data structure design, coding, cloud infrastructure work, DevOps, and more to implement some or all of the below proposed features:
- Expand data sets
- Add additional states’ data
- Design and implement more advanced filtering
- Design and implement more advanced visualization
- Create new strategies and implementations for data-mining / analytics / AI / new algorithms