Skip to main
Binary code

Three CDS students honored for their poster submissions to the 10th Annual College of Science Undergraduate Research Colloquium

In Physical Sciences and Mathematics  (Physics, Chemistry, Astronomy, Mathematics, Computer Science), Steven Hoang presented his award-winning poster on "A Recommendation System for e-Learning Videos on YouTube."   Mentors: Ron Mahabir and Olga Gkountouna.

Abstract:  Video sharing platforms allow users seeking mastery in a professional skill or topic to access a widespread range of educational content using specific keywords. However, for users with limited domain knowledge of the specific field, or unsure of what keywords to use, this can lead to poor recommendations. This project aims to build a recommendation engine that takes as input users’ professional goals and returns video content most relevant to them. The LDA algorithm is used to build a model of ranked topics from computer science job postings. The trained model is then applied to the transcripts of YouTube videos. A developed front-end interface allow users to input their professional objectives and preferences, which is classified by the model. A collection of related job postings is then returned to the user to evaluate their relevance. Relevant job postings are then compared to transcripts of the same topic, and videos with the highest similarity are recommended to the user.

Kamryn Cullen presented her award-winning poster on “Machine Learning for the Forecasting of Alzheimer’s Disease Clinical Status.”  Mentors: Ron Mahabir and Olga Gkountouna.

Abstract: Alzheimer’s Disease is a neurodegenerative disease that affects the brain and is commonly seen in older patients. The use of Machine Learning to predict Alzheimer’s has been the focus of research in recent years and has promising potential to help diagnose and forecast patient health status based on clinical data. Alzheimer’s Disease is unique in that there are many complex heterogeneous factors that play a part in its progression, and which are often overlooked or undiagnosed in its early stages. Having accurate models to predict the progression of this disease assists with treatment options and supports the possibility of early detection. This project will use publicly accessible clinical data to train machine learning classification models to predict patients' disease status. Our work provides valuable insights into the various progressive and confirmed determinants of Alzheimer’s Disease.

In Life Sciences (Biology, Neuroscience, Forensic Science), Oliver Yu, Computational and Data Sciences, presented his Geospatial Mapping of Human Tissues to Advance Precision Medicine for Cancer Patients (Poster #13) . Mentor: Mariaelena Pierobon.