10 Summary

10.1 Topics covered

10.1.1 Exploratory data analysis

1. Data Wrangling
   • Subsetting, creating variables, reshaping, and summarizing
   • data.table
   • dplyr
   • sparklyr and Apache Spark
2. Data Visualization
   • Static visualizations
   • ggplot2
   • ggplot2 add-ons GGally, ggExtra, and ggalluvial
   • cowplot as an additional ggplot2 theme
   • Interactive visualizations
   • plotly

10.1.2 Supervised learning

3. Decision trees
   • Classification and regression trees
   • rpart
   • rpart.plot
4. Random forests
   • Random forests for classification and regression
   • randomForest
5. Model building and evaluation
   • modelr
   • caret
6. Support Vector Machines
   • Maximal margin classifers
   • Support vector classifiers
   • Support vector machines with polynomial and radial kernels
   • Logistic regression
   • Tuning and evaluating the models
   • e1071
   • sparklyr

10.2 Methods we didn’t cover

1. Regression
   • Penalized regression
     • ridge, lasso, elastic net
     • glmnet
   • Principal components and partial least squares (a supervised version of PC) regression
     • pls
   • Non-linear regression
     • Polynomials, splines (smoothing splines), generalized additive models
     • splines
     • gam
2. K-nearest-neighbors (KNN)
   • caret
   • class
3. Unsupervised learning
   • K-means clustering
   • Hierarchical clustering
   • cluster
   • factoextra