Alley Heaps student internship helps shape academic journey, contribute to impactful research

October 1, 2024

Pictured, l-r, are Caelen Mattie, Kennedy Roland, and Alastair May. Missing is AJ Hinman.

Transformative is how one student describes his research experience this summer as a recipient of the Alley Heaps Undergraduate Research Internship, awarded to StFX computer science students. The research opened a door he had never considered before.

Other students say their experience was invaluable, learning from a faculty member, and contributing to research that will positively impact society.

This year, students AJ Hinman, Caelen Mattie, Alastair May, and Kennedy Roland each received $8,500 for summer research work under the supervision of a StFX faculty member. The internships are part of the Dr. H. Stanley & Doreen Alley Heaps Chair, which provides for the support, exploration, and advancement of computing science at StFX.

SHAPED ACADEMIC JOURNEY

"This research opportunity has been transformative, particularly in how it has shaped the direction of my academic journey," says AJ Hinman of Turner Valley, AB, a third year computer science student who is also completing a Diploma in Engineering.

Working with Dr. Milton King, he conducted research that involved using a language model to explore the linguistic differences in social media communications, based on the skills users list on their LinkedIn profiles.

"The primary focus was on comparing how individuals with technical backgrounds, such as engineers or programmers, communicate differently from those without such backgrounds. The goal was to better understand the relationship between professional skills and online language patterns, which could have implications for personalized content delivery and user engagement strategies."

By diving into Natural Language Processing (NLP), he's been able to explore the intersection of language and technology, an area that he hadn't fully considered before.

"This experience has not only deepened my understanding of NLP but has also influenced me to consider specializing further in this field as I continue my studies. The research has opened a door I hadn't considered opening, leading me to investigate grad school in computer science."

This experience provided him with a blend of technical and analytical skills, particularly in working with NLP models like BERT. "The research has deepened my knowledge of Python, showing me how to access more information for personal and professional projects. It has also given me valuable insight into how research is conducted, from hypothesis formation to data analysis and interpretation. These skills are directly applicable to my future career and have made me more confident in pursuing complex, interdisciplinary projects."

What stands out most, he says, is the hands-on application of theoretical knowledge to real-world problems, particularly through working with large datasets and sophisticated models.

INSIGHTS FROM EXPERTS

Kennedy Roland of St. Catharines, ON, now in her final year in computer science, researched the effects of time of day on next-word prediction this summer. "I looked at large language models like GPT and trained them using 3hr time segments to see if they can better predict what word is going to be typed next in comparison to when they are trained randomly or on the author's other texts (at different times)," she says.

Ms. Roland, who was supervised by Dr. Milton King, says the project was a fantastic opportunity to explore a new-to-her field in computer science. The research award also enabled her to gain insights from her supervisor, an expert in the field, which made it that much more beneficial.

"The award allowed me to dedicate a lot of time to learning and playing around with the code and getting everything to make sense. I otherwise would not have had the time to pursue this interest in this much detail and would have missed out on getting the experience of working on something like this."

She says the experience really prepared her for her honours thesis this year. "I gained a lot of valuable research experience that is specific to computer science, as well as experience working with a supervisor and having regular check-ins and progress updates on a large-scale project, something that is typically left up to the student for any normal school project. I think this is a great experience to speak to in job interviews and have on my resume as I can show some detailed work that came along from it."

She says the opportunity to work with a professor is so valuable. "You get a lot of mentoring and can achieve a lot from their guidance while still doing a completely independent project."

REWARDING TO SEE WORK HELP OTHERS

Caelen Mattie of Antigonish, NS, a fourth year student completing an advanced major in computer science with business, worked on a project using machine learning methods to better capture human emotion from audio samples.

"Using supervised shallow neural networks, we found that, given enough data, a model can generalize and extract paralinguistic information from audio for emotion recognition purposes. The research involved a literature review of the speech emotion recognition field, analysis of available datasets, and the creation of a python pipeline which maps audio samples to labels in an attempt to classify samples of an unseen emotion," he says.

"I utilized many tools such as pytorch, tensorflow, optuna, fasttext, sklearn, Nvidia CUDA and others."

Mr. Mattie, who was supervised by Dr. Milton King, says it's been rewarding to see his work go into creating something that could be useful to other people. "The field of speech emotion recognition will impact many areas of our everyday life from customer service to healthcare. I appreciate the opportunity to work on this project and learn from our talented faculty here at StFX."

He says this experience has given him valuable insight in the field of artificial intelligence and machine learning which will be important fields in the future. "Computer science research is growing rapidly and I feel this experience has given me skills that are useful and will apply to my future research."

Collaborating with other students working on similar problems and hearing about what they have done during their research was also a highlight.

Alastair May worked under the supervision of Dr. Taylor Smith on "Improving Student Learning via Interactive AI."