Melissa R. Dale, PhD
Machine Learning Research & Data Scientist
Modeling complex systems under uncertainty
ABOUT
I’m a machine learning researcher with a PhD in Computer Science and Engineering, and I’m drawn to problems where the data is messy, incomplete, or difficult to interpret. I build practical, carefully evaluated models that combine information from multiple sources, with a strong focus on understanding when a model works, when it fails, and what its results actually mean. I enjoy turning complex technical questions into clear, reproducible analyses that can support better real-world decisions.
EDUCATION
Michigan State University
2016 – 2026
PhD: Computer Science and Engineering
Information Fusion, Machine Learning, AutoML, Computer Vision
Montana State University
2011 – 2014
Master of Science: Computer Science
Montana State University
2006 – 2011
Undergraduate, Magna Cum Laude
BS in Computer Science
BA in Modern Languages (Spanish)
Minor in Physics
PUBLICATIONS
To Impute or Not: Recommendations for Multibiometric Fusion
WIFS 2023
Dale M., Singer E., Borgström B., Ross A.
IEEE International Workshop on Information Forensics and Security, Germany, December 2023.
On the Design of the MIT LL Trimodal Dataset for Identity Verification
IWBF 2023
Singer E., Borgström B. J., Alperin K., Nguyen T., Dagli C., Dale M., & Ross A.
IEEE 11th International Workshop on Biometrics and Forensics, Barcelona, April 2023.
Addressing Missing Scores in Evolving Multibiometric Systems
ICPR 2022
Dale M., Jain A., Ross A.
26th International Conference on Pattern Recognition, Canada, August 2022.
Fusing AutoML Models: A Case Study in Medical Image Classification
ICPRAI 2022
Dale M., Ross A., Shapiro E.
3rd International Conference on Pattern Recognition and Artificial Intelligence, France, June 2022.
Impacts of Design Pattern Decay on System Quality
ESEM 2014
Dale M., Izurieta C.
8th ACM-IEEE International Symposium on Empirical Software Engineering and Measurement, Torino, September 2014.
EXPERIENCE
Machine Learning Researcher
2016 – Present
iPRoBe Lab @ Michigan State University
I design and evaluate machine learning systems that integrate multiple data sources to improve
predictive performance under uncertainty. My work focuses on understanding when additional data
improves predictions and how to build robust models across heterogeneous datasets.
Technologies: Python, Scikit-learn, Pandas, NumPy, Matplotlib, AutoML, R
Teaching Assistant, Instructor
2014 – 2020
Montana State University, Michigan State University
Taught and supported courses in programming, algorithms, data structures, biometrics,
discrete mathematics, web design, and ethics in computing.
Software Engineering Research Assistant
2012 – 2016
Software Engineering Lab @ Montana State University
Developed analytical tools to study software architecture, design pattern degradation,
technical debt, and long-term maintainability in large codebases.
Technologies: Java, R
XBRL Services Software Engineer Intern
2013 – 2014
WebFilings / Workiva
Refactored ~10,000 lines of legacy code across 30 files into a consolidated, maintainable Python module, improving code clarity and simplifying future development
Technologies: XBRL, Java, Python
PROJECTS
Multimodal Deception Detection
2026 – Present
Developing a reproducible machine learning pipeline for detecting deception from
narrative text and behavioral typing signals. Built data validation and feature
extraction tools, fixed participant-level train/test and cross-validation splits,
and classical NLP baselines designed to prevent participant leakage.
Technologies: Python, Pandas, NumPy, Scikit-learn, NLP, TF-IDF, Jupyter
Postie Data Science Case Study
2026
Analyzed transaction-level e-commerce data to identify normalization problems,
outliers, unexpected values, and reporting pitfalls. Investigated purchasing
patterns and product information, evaluated meaningful business metrics, and
produced an uncertainty-aware sales forecast with clearly documented recommendations.
Technologies: Python, Pandas, NumPy, Scikit-learn, Matplotlib, Jupyter
Score Fusion App
2020
Developed an interactive framework for ingesting, normalizing, combining, and
evaluating prediction scores from multiple models and data sources. The application
supports strategy comparison, performance analysis, and visualization across
biometric and other score-based classification problems.
Technologies: Python, Kivy, Pandas, Scikit-learn, Matplotlib
Map Generation Pipeline
2025 – Present
Built a configurable geospatial pipeline for extracting, processing, and rendering
OpenStreetMap data. Developed reusable workflows for selecting geographic features,
transforming spatial data, and generating large-scale custom map visualizations.
Technologies: Python, OSMnx, GeoPandas, Matplotlib
Grime Injector
2014
Java-based research tool for simulating architectural decay and measuring the impact
of design pattern erosion on software quality.
Technologies: Java, SonarQube, JAD, Javassist
BEYOND THE RESUME
I care about mentoring the next generation of STEM students and making technical material more approachable. Outside of research, I enjoy creative technical projects that combine programming, design, and real-world problem solving.