Summary

In this chapter, you have learned the importance of data preparation in predictive modeling, which involves both data exploration and data visualization exercises. R has a number of open-source packages that are useful for data munging, for example dplyr, reshape, and many more. The challenge is to hit the right balance between having data munging activities in SQL Server VS in R. The beauty of SQL Server Machine Learning Services is that it allows easy integration with SQL Server Reporting Services. In addition, Power BI also supports interactive data exploration with R visualizations. In the next chapter, you will learn more about the RevoScaleR library for portable, scalable, and distributable R functions.