Predictive analytics delivers the greatest value when the data being modeled is relevant to the business goals. Explore the preprocessing phase of data collection to provide the best predictive model.
- start the course
- recognize what is tidy and what is untidy data
- identify outliers and determine whether to remove these values
- perform data transformation, normalization, and scaling
- recognize important aspects of variable partitioning
- recognize important aspects of setting dummy variables and removing variables
- recognize key approaches for handling missing data
- use imputation to replace missing data
If you would like to provide feedback for this course, please e-mail the NICCS SO at NICCS@hq.dhs.gov.