Purpose of the book

This is not a technical guide about R and Qlik Sense integration, or a Rattle guide for software developers. This book is an introduction to the basic techniques of predictive analytics and data visualization. We've written this book for business analysts, and people with an IT background, but without analytics experience.

We believe that the best way to learn is by practicing, and for this reason this book is organized around examples, which you can do with a simple Windows computer. Don't be afraid, we will use two software tools, Rattle and Qlik Sense Desktop, in order to avoid complex code. To create the predictive analysis, we'll use Rattle and for data visualization, we'll use Qlik Sense Desktop.

There are two ways of using Rattle, or R, and Qlik Sense Desktop together. These are listed as follows:

• In the first approach, it is possible to integrate Qlik Sense Desktop and R. The business users select some data. Qlik Sense Desktop sends this selected data to an R server, the server processes the data and performs a prediction. The R server returns the data to Qlik Sense Desktop, and this shows the data to the user. This model has a great advantage—the interactivity, but it also has a disadvantage; it requires additional software to integrate the two different environments.
• The second approach is based on two steps. In the first step, the R environment loads the data, performs the prediction, and stores the original data with the prediction. In the second step, Qlik Sense Desktop loads the data and the prediction, and shows it to the business user. This second approach has a great advantage which is simplicity, but also has a disadvantage which is the lack of interactivity.

In this book, we'll use the second approach because in predictive analytics choosing the appropriate model is the key. For this reason we want to focus on introducing you to different models, avoiding the technical stuff of integration. We'll use Rattle and Qlik Sense Desktop in a two-step process. We'll load data in Rattle to enrich it with a predictive model and then load it in Qlik Sense Desktop to share it by creating data visualizations. This process is illustrated in the following diagram: