You are previewing Predictive Analytics Using Rattle and Qlik Sense.
O'Reilly logo
Predictive Analytics Using Rattle and Qlik Sense

Book Description

Create comprehensive solutions for predictive analysis using Rattle and share them with Qlik Sense

In Detail

Qlik Sense Desktop, the personal and free version of Qlik Sense, is a powerful tool for business analysts to analyze data and create useful data applications. Rattle, developed in R, is a GUI used for data mining and complements Qlik Sense Desktop very well. By combining Rattle and Qlik Sense Desktop, a business user can learn how to apply predictive analytics to create real-world data applications. The objective is to use Qlik Sense to analyze data and complement it with predictive analytics using Rattle.

This book will introduce you to basic predictive analysis techniques using Rattle and basic data visualizations concepts using Qlik Sense Desktop. You will start by setting up Qlik Sense Desktop, R, and Rattle and learn the basic of these tools. Then this book will examine the data and make it ready to be analyzed. After that, you will get to know the key concepts of predictive analytics, by building simple models with Rattle and creating visualizations with Qlik Sense Desktop. Finally, the book will show you the basics of data visualization and will help you to create your first data application and dashboard.

What You Will Learn

  • Set up your desktop environment by installing Qlik Sense Desktop, R, and Rattle

  • Explore Rattle charts and the most commonly used multivariate statistical techniques to discover relationships among data

  • Find solutions to business questions by applying data analysis techniques

  • Use unsupervised and supervised learning methods to gain insights into your data

  • Evaluate the performance of a predictive model

  • Create basic charts and filters using Qlik Sense Desktop to build your first data application

  • Improve your analysis by complementing Qlik Sense Desktop with predictive analytics

  • Familiarize yourself with the basics of data visualization and data storytelling

  • Downloading the example code for this book. You can download the example code files for all Packt books you have purchased from your account at If you purchased this book elsewhere, you can visit and register to have the files e-mailed directly to you.

    Table of Contents

    1. Predictive Analytics Using Rattle and Qlik Sense
      1. Table of Contents
      2. Predictive Analytics Using Rattle and Qlik Sense
      3. Credits
      4. About the Author
      5. About the Reviewers
        1. Support files, eBooks, discount offers, and more
          1. Why subscribe?
          2. Free access for Packt account holders
          3. Instant updates on new Packt books
      7. Preface
        1. What this book covers
        2. What you need for this book
        3. Who this book is for
        4. Conventions
        5. Reader feedback
        6. Customer support
          1. Downloading the example code
          2. Downloading the color images of this book
          3. Errata
          4. Piracy
          5. Questions
      8. 1. Getting Ready with Predictive Analytics
        1. Analytics, predictive analytics, and data visualization
        2. Purpose of the book
        3. Introducing R, Rattle, and Qlik Sense Desktop
        4. Installing the environment
          1. Downloading and installing R
          2. Starting the R Console to test your R installation
        5. Downloading and installing Rattle
        6. Installing Qlik Sense Desktop
        7. Exploring Qlik Sense Desktop
        8. Further learning
        9. Summary
      9. 2. Preparing Your Data
        1. Datasets, observations, and variables
        2. Loading data
          1. Loading a CSV File
        3. Transforming data
          1. Transforming data with Rattle
            1. Rescaling data
            2. Using the Impute option to deal with missing values
            3. Recoding variables
            4. Binning
          2. Indicator variables
            1. Join Categories
            2. As Category
            3. As Numeric
        4. Cleaning up
        5. Exporting data
        6. Further learning
        7. Summary
      10. 3. Exploring and Understanding Your Data
        1. Text summaries
          1. Summary reports
            1. Measures of central tendency – mean, median, and mode
            2. Measures of dispersion – range, quartiles, variance, and standard deviation
              1. Range
              2. Quartiles
              3. Variance
              4. Standard deviation
            3. Measures of the shape of the distribution – skewness and kurtosis
          2. Showing missing values
        2. Visualizing distributions
          1. Numeric variables
            1. Box plots
            2. Histograms
            3. Cumulative plots
          2. Categorical variables
            1. Bar plots
            2. Mosaic plots
        3. Correlations among input variables
          1. The Explore Missing and Hierarchical options
        4. Further learning
        5. Summary
      11. 4. Creating Your First Qlik Sense Application
        1. Customer segmentation and customer buying behavior
        2. Loading data and creating a data model
          1. Preparing the data
        3. Creating a simple data app
        4. Associative logic
        5. Creating charts
        6. Analyzing your data
        7. Further learning
        8. Summary
      12. 5. Clustering and Other Unsupervised Learning Methods
        1. Machine learning – unsupervised and supervised learning
          1. Cluster analysis
            1. Centroid-based clustering the using K-means algorithm
            2. Customer segmentation with K-means clustering
              1. Preparing the data in Qlik Sense
            3. Creating a customer segmentation sheet in Qlik Sense
          2. Hierarchical clustering
          3. Association analysis
        2. Further learning
        3. Summary
      13. 6. Decision Trees and Other Supervised Learning Methods
        1. Partitioning datasets and model optimization
        2. Decision Tree Learning
        3. Entropy and information gain
        4. Underfitting and overfitting
        5. Using a Decision Tree to classify credit risks
          1. Using Rattle to score new loan applications
          2. Creating a Qlik Sense application to predict credit risks
        6. Ensemble classifiers
          1. Boosting
          2. Random Forest
          3. Supported Vector Machines
        7. Other models
          1. Linear and Logistic Regression
          2. Neural Networks
        8. Further learning
        9. Summary
      14. 7. Model Evaluation
        1. Cross-validation
        2. Regression performance
          1. Predicted versus Observed Plot
        3. Measuring the performance of classifiers
          1. Confusion matrix, accuracy, sensitivity, and specificity
          2. Risk Chart
          3. ROC Curve
        4. Further learning
        5. Summary
      15. 8. Visualizations, Data Applications, Dashboards, and Data Storytelling
        1. Data visualization in Qlik Sense
          1. Visualization toolbox
          2. Creating a bar chart
            1. The Data menu
            2. The Sorting menu
            3. The Add-ons menu
            4. The Appearance menu
        2. Data analysis, data applications, and dashboards
          1. Qlik Sense data analysis
            1. In-memory analysis
            2. Associative experience
          2. Data applications and dashboards
            1. The DAR approach
        3. Data storytelling with Qlik Sense
          1. Creating a new story
        4. Further learning
        5. Summary
      16. 9. Developing a Complete Application
        1. Understanding the bike rental problem
        2. Exploring the data with Qlik Sense
          1. Checking for temporal patterns
          2. Visual correlation analysis
        3. Creating a Qlik Sense App to control the activity
        4. Using Rattle to forecast the demand
          1. Correlation Analysis with Rattle
          2. Building a model
          3. Improving performance
        5. Model evaluation
          1. Scoring new data
        6. Further learning
        7. Summary
      17. Index