Dr. Allison Theobold

Assistant Professor, Statistics Department

Cal Poly in San Luis Obispo 

This introductory workshop will cover the following topics:

• Cleaning data

• Visualizing relationships

• Choosing predictor (input) variables

• Choosing between competing models

• Interpreting model conclusions

• Data ethics

The workshop will be offered in both R and Python, so scholars can choose which language they prefer. Ideally, participants will be able to work locally (on their computer) with a Quarto file (.qmd) or a Python notebook (.ipynb).

Dr. Wasila Dahdul

Data Curation Librarian, UCI Libraries

University of California, Irvine

Secondary data is data originally collected by others and available for reuse. Such data can support large-scale analyses, enable new research directions, and provide access to data that is difficult or impossible to collect firsthand. This workshop will provide an overview of data sources and data discovery platforms for a variety of disciplines. We will then consider key issues in reusing data by using R to explore and visualize a real-world dataset, the U.S. National Parks Visit Data. Working through this example, participants will assess data suitability and quality, and gain an understanding of how to use documentation, provenance, and the social and historical context to work responsibly with data. 

Dr. Kelly Bodwin

Associate Professor, Statistics Department

Cal Poly in San Luis Obispo 

In this workshop, we will deep dive into three major elements of the machine learning workflow: model specification, feature engineering, and model tuning.  Using real world example data, we will assess which of these three has the biggest impact on a model's predictive success.  We'll also discuss strategies for choosing a feasible number of pipeline variants to test, and how to iterate progressively in the model fitting process.

This material assumes the learner is familiar with the basic pipeline structure of Scikit-learn (python) or tidymodels (R); with common variable transformations like logarithms or interaction terms; and with common assessment metrics such as MSE, R-squared, MAE, accuracy, precision, and recall. 

Dr. Wasila Dahdul

Data Curation Librarian, UCI Libraries

University of California, Irvine

Secondary data is data originally collected by others and available for reuse. Such data can support large-scale analyses, enable new research directions, and provide access to data that is difficult or impossible to collect firsthand. This workshop will provide an overview of data sources and data discovery platforms for a variety of disciplines. We will then consider key issues in reusing data by using R to explore and visualize a real-world dataset, the U.S. National Parks Visit Data. Working through this example, participants will assess data suitability and quality, and gain an understanding of how to use documentation, provenance, and the social and historical context to work responsibly with data. 

Brian Vegetabile is a Senior Data Scientist and Applied Researcher at LinkedIn. He is an expert in Causal Inference and Machine Learning and is currently working in the Product Services & Marketplace Data Science organization focused on Systems and Marketplace Understanding. Prior to his current team, he worked on problems related to forecasting, anomaly detection, and root cause analysis focused on applied operations problems. He earned his bachelor's in Aerospace Engineering at Penn State before his PhD and Master's in Statistics at UC Irvine. 

Dr. Nuño received a B.S. in Applied Mathematics from the University of California, Riverside in 2015, then went on to complete graduate work at the University of California, Irvine, where she received an M.S in Statistics in 2017 and a Ph.D. in Statistics in 2020. Her research interests include clinical trials and the development of robust statistical methodology for efficient sampling designs. She is particularly interested in biomedical applications in Alzheimer’s disease and pediatric oncology. 

After graduating with a B.S. in Statistics from CSU Monterey Bay in 2022, Sophia pursued a career as a data analyst in the technology sector. As an individual contributor at ConnectWise, her priorities were system implementation, churn prediction and analysis, and data pipeline management. She now leads a team of analysts focused on integrating technology and people, improving how internal teams interact with technological systems, processes, and automations.