No matter what you call it – data munging, data cleansing or data wrangling – everyone agrees that data preparation activities account for 80 percent of analysts’ time, leaving only 20 percent for analysis. Shifting this work to more specialized talent represents a major source of data analysis productivity improvements.
In this session, Peter Aiken, Professor of Information Systems at Virginia Commonwealth University, will discuss the major preparation categories, including collection, evaluation, evolution, access design and storage requirements. Understanding each in context also provides opportunities to develop data governance/ethics frameworks.
You’ll learn how to:
- Transform data preparation from one-off to a more efficient, time-saving process
- Identify and fill data preparation gaps
- Understand the transformations data can survive as it’s prepared for analysis