Predictive Modeling & Analytics

This course takes a problem-based learning approach to the foundational concepts of data science. Program participants will learn the process of developing and completing a data science project by determining which tools to use to solve analytic problems. It is designed for individuals with limited to no background in data science, although some basic knowledge of mathematical and statistical processes would be helpful. Some programming experience is highly recommended. Successful completion will contribute to the DataPro III Series. 

Key Takeaways

  • Develop and complete a real-world data science problem using best-practice tools, analytics, and solutions
  • Confidently understand and evaluate predictive modeling, data scaling,
    normalization, and their uses
  • Learn to select the best models and analytics for your problem
  • Make better decisions using quantifiable, data-driven evidence
  • Leverage your data science fluency for career development, enhanced skill sets, internal and external marketability
  • Apply the tools and language of predictive modeling and analytics to your current position and organization
  • Personalize your learning with a facilitated, blended experience


  • Building a Predictive Model: End-to-End Perspective (3 units)
  • Integrating Data Scaling and Normalization into End-to-End Model Building (3 units)
  • Handling Categorical Data as Preparatory Data (3 units)
  • Sense Making: Evaluating and Validating Predictive Models (3.5 units)
  • Selecting the Approach: Linear-Logistic-Multinomial-Poisson-Ordinal Modeling (3.5 units)
  • Optimizing the Models: Ensembles and Hyperparameter Tuning (3.5 units)

Who Should Attend

  • Early to mid-career professionals interested in math, science, and coding who want to improve their internal or external marketability
  • IT professionals looking to pivot or advance their career track
  • Senior managers and executives charged with leading one or more data scientists and incorporating data science or machine learning into their operations
  • Mid-level professionals who engage with large amounts of external data, such as market information
  • Functional leaders seeking confidence to engage in intelligent conversations about data science


Nagiza Samatova

Dr. Samatova is a Professor of Computer Science at NC State University, specializing in Big Data analytics, Graph Theory, Graph Algorithms and Graph Mining, and Data-Drive Discovery from Scientific Data. One of her hallmarks is pioneering work on parallel R. Her broad impact has been read into the Congressional records (2006), recognized by The New York Times and Science magazine (2007, 2010) and widely acclaimed throughout her field.

Learn More About Nagiza Samatova

Request more information