Open to Research and Consulting Projects
I'm Justin Philip Tuazon, a data scientist and researcher working in statistics and machine learning. I develop and study quantitative methods, and apply them to challenging problems.
Galton board: random bounces settling into a normal curve
I recently presented our research, Pairwise Target Rotation for Factor Models, in an oral session at the 2026 International Meeting of the Psychometric Society (IMPS) held at Yonsei University in Seoul, South Korea back in July 2026. This work focuses on developing an intuitive yet flexible method for incorporating a priori information, such as item semantics, into factor rotations. The approach can be used to produce interpretable factor solutions and effectively characterize latent variables.
Our work, Design and Implementation of Reduced Longitudinal Experiments, was selected from over 13,000 submissions for poster presentation at the 2026 American Educational Research Association (AERA) Annual Meeting. It focuses on characterizing hierarchical linear models to produce practical recommendations for designing and implementing longitudinal experiments. My co-author, Le Dolino, presented our paper at the conference back in April 2026 in Los Angeles, California.
My tech stack
Python
R
AWS SageMaker
Snowflake
Tableau
Power BI
Streamlit
SharePoint
LaTeXProfile & History
I am an early-career Data Scientist and Researcher passionate about statistics and machine learning. My work centers on developing novel quantitative methods, as well as intelligent systems, and creatively applying such, ultimately combining rigorous modeling with clever approaches to uncover complex patterns in data. Currently, my research and professional practice focus on latent variable models, mixed-effects models, and causal machine learning.
Moreover, I am currently pursuing an M.S. in Computer Science degree at the University of the Philippines Diliman (UPD), where I am affiliated with the Computer Vision and Machine Intelligence Group of the Department of Computer Science. I previously earned my B.S. in Statistics degree from UPD, graduating summa cum laude back in 2023.
Outside of academics and work, I enjoy playing video games and playing the guitar. 🙂
Selected Works
Click on the icon(s) at the top right of each card for more details.
We developed a new factor rotation method (an unsupervised learning technique) and its Python implementation, that can incorporate arbitrary prior information (e.g., semantics) for exploration and model building.
Presented at the 2026 International Meeting of the Psychometric Society (IMPS) in an oral session.
Won 1st Place Best Graduate Student Paper Award at 25th Student-Faculty Conference on Statistical Sciences.
We characterized hierarchical linear models through the design and analysis of a Monte Carlo experiment (including the simulation of nested data, attrition, and more), implemented in R, to produce recommendations for longitudinal experiments.
Presented at the 2026 American Educational Research Association (AERA) Annual Meeting in a poster session.
I developed an LLM-enhanced visual workbench for exploratory factor analysis with interactive model building and automated interpretation.
I implemented Gower’s Distance and Binary Integer Linear Programming using R and created a web app for a student council initiative that involved optimally grouping students for theses.
I applied computer vision methods and attention-based multiple instance learning to predict the isocitrate dehydrogenase (IDH) mutation status of a glioma based on magnetic resonance imaging (MRI) scans, exploring a non-invasive way of classifying gliomas while including an ablation study that combines scans with radiomics.
I made a game for one of my courses during my first semester in college. The game was made using Unity, with C# as the programming language. The simple design assets were made using Microsoft PowerPoint while the background music was created using Chrome Music Lab - Song Maker.
Technical Capabilities
Experienced with core machine learning (e.g., supervised and unsupervised learning, classification and regression, recommender systems), statistical modeling (e.g., mixed-effects models, generalized linear models, factor models, uplift models), experimental design, hypothesis testing, computational statistics, and applied domains (e.g., computer vision, natural language processing).
Achievements & Honors
Engagements
During the inaugural UnionBank Negosyante Boostcamp, I served as a resource speaker for a session on Artificial Intelligence in Business. I introduced participants to AI fundamentals, including its definitions, types, and practical business use cases. Through practical examples and discussions, I helped participants understand how AI can address business challenges, improve productivity, and create new opportunities for growth.
During the Data Science and Analytics Group Q2 Townhall at UnionBank of the Philippines, I gave a Knowledge Sharing session (i.e., lecture) on latent variable modeling (particularly, factor models). During the talk, I provided an introduction to factor analysis, including its mathematical background, its applications, and some recent developments in the field.
I gave a talk to over 30 young students and aspiring Pisay scholars, sharing my journey from school to a career in STEM. I provided advice, answered questions, and hopefully inspired the next generation of Filipino innovators to embrace the challenges ahead.
I co-facilitated a research workshop for students and educators, collaborating with colleagues to instruct participants on the end-to-end research process, technical concepts, and quantitative methodologies.
A quick game: how well can you read a scatterplot?
How strongly are these two variables related? Guess Pearson's r, from −1 to 1.
A streak counts guesses within 0.10 of the actual r.
Your guesses against the actual r. Perfect guesses sit on the dashed line.
Interested in statistics, machine learning, and data science? Feel free to connect!