Data construction is arguablythe least understood of the five major data preparation tasks for analytics. This half-day vendor-neutral session will expose analytics practitioners, data scientists, and those looking to get started in predictive analytics to the critical importance of properly assembling data in advance of model building. This deep dive into the topic will provide several step-by-stepexamples of the process.
The instructor will show a variety of examples starting with data exploration and preliminary modeling. Various possible calculations and transformations will be considered and demonstrated. The instructor will then explore which of the various options seem most promising for modeling. Attendees will learnhow seeminglyindependent ingredients can make a far more potent mix for analytics insight than the original raw variables were ever capable of.
You Will Learn
- Why proper data construction is so important for analytics impact
- The iterative process from exploration to experimentation to confirmation
- he implications of data construction on model deployment
- The classic traits that appear in most data setsand how to treat them
- Effective techniquesfor improving data’sability to make predictions
- Analytics Practitioners; Data Scientists; IT Professionals; Technology Planners; Consultants; Business Analysts; Analytics Project Leaders