Spend some time with us this fall and learn new tools and techniques from our internal data science experts. We hope that you are able to join us for one or more of our data science webinars. See details below and be sure to register by filling out the form at the bottom of each description.

SQL for Data Scientists

December 2, 2021 
2:00 pm - 3:00 pm Eastern
Presenter: Renee Teate, Director of Data Science, HelioCampus

We are excited to announce that HelioCampus’ own Director of Data Science, Renee Teate, has written a book: SQL for Data Scientists: A Beginner's Guide for Building Datasets for Analysis. At HelioCampus, Renee uses her SQL skills to build datasets for predictive models, helping our higher education institutions to predict enrollment trends, package financial aid and more.

In this webinar, she’ll share more about the role of SQL in her work, specifically how it helps transform the data from higher ed institutions into our standard data extracts and then is used to prepare the data for our predictive models. If you are interested in learning more about SQL, Renee’s journey to becoming a data scientist, why she wrote the book, an open Q&A and more, you won’t want to miss this webinar. We hope to see you online!


Time Series Methods for Enrollment Forecasting

December 16, 2021
2:00 pm - 3:00 pm Eastern
Presenter: Karen Heil, Data Scientist, HelioCampus

For higher education institutions, forecasting future enrollments is key for budget and resource planning. Naïve methods relying on historical conversion metrics and upstream indicators can provide estimates of expected outcomes. However, such methods are not responsive to early signals and do not naturally account for year to year variation in patterns.

Time series forecasting can provide actionable insights based on emerging trends and, when correctly calibrated, generate strong predictions even as conditions on the ground change. This webinar will discuss the requirements for generating a valid forecast, including ways to overcome some common challenges, and walk through two recent case studies demonstrating how well calibrated forecasts are able to accurately predict enrollments despite significant changes in procedures and processes over the last two years at two different institutions.


Predictive Modeling Project Success Through Conversation, Process, and Transparency

January 13, 2022
2:00 pm - 3:00 pm Eastern
Presenter: Rick Ruiz, Senior Data Scientist, HelioCampus

Predictive models are sometimes delivered as black boxes treated as if they're imbued with magic data science pixie dust. This lack of transparency makes it difficult to understand and mitigate the effects of changing conditions, like we’ve seen during the pandemic, or to determine whether the model is performing poorly for a subset of the population, potentially perpetuating biases.

An alternative is investing in predictive modeling as a process that manifests itself as an ongoing conversation between subject matter experts at each institution and the data scientists training the models, with a focus on delivering a model that can be used to inform both policies and interventions to improve individual student outcomes. Today’s presentation will walk you through the contours of this conversation and help set expectations for predictive modeling, while introducing helpful concepts you can use to orient yourself and your colleagues when navigating your next predictive modeling project.


Tools for Evaluating and Explaining Predictive Models

January 27, 2022
2:00 pm - 3:00 pm Eastern
Presenter: Brian Richards, Data Scientist, HelioCampus

At HelioCampus, we go to extra lengths to ensure that our predictive models are as transparent and explainable as possible. We train multiple models using different input fields, algorithms, and settings to ensure we’re producing predictions that make sense and are useful for answering the question at hand. We’ve recently implemented a new set of tools that allow us to more easily and effectively compare model performance and assess which input features have the most influence on each student’s score. Used in the aggregate, these tools can also help identify areas of concern that may be addressed at the policy level.

In this webinar, we’ll highlight the tools our Data Science team uses for model evaluation and explainability, how we use them internally to check our model results, and what we provide to our clients to help increase understanding and adoption of our predictions.