1. Machine Learning in R and its use in the statistical offices
stat.unido.org

2. Outline Machine learning and R R packages
Machine learning in official statistics
Top 10 algorithms
References

3. What I talk about when I talk about Machine Learning3

4. R and R packages What makes R so useful?
The users can extend and improve the software or write variations for specific tasks.
The R package mechanism allows packages written for R to add advanced algorithms, graphs, machine learning and and mining techniques
Each R package provides a structured standard documentation including code application examples

5. R and R packages ## Naive Bayes example
> install.packages('e1071', dependencies = TRUE)
> library(class)
> library(e1071)
> data(iris)
> pairs(iris[1:4], main = "Iris Data (red=setosa,green=versicolor,blue=virginica)", pch = 21, bg = c("red", "green3", "blue")[unclass(iris$Species)])

7. R and R packages
> classifier <- naiveBayes(iris[,1:4], iris[,5])
> table(predict(classifier, iris[,-5]), iris[,5])
setosa versicolor virginica
setosa
versicolor
virginica

8. Machine Learning for Official Statistics
Automatic Coding
Editing and Imputation
Record Linkage
Other Methods

9. Automatic coding Automatic coding via Bayesian classifier: caret, klaR
Automatic occupation coding via CASCOT: algorithm not described
Automatic coding via open-source indexing utility: ?
Automatic coding of census variables via SVM: e1071 (interface to libsvm)

10. Editing and Imputation
Categorical data imputation via neural networks and Bayesian networks: neuralnet, gRain, bnlearn, deal
Identification of error-containing records via classification trees: rpart, tree, caret
Imputation donor pool screening via cluster analysis: class, klaR, cluster, kmeans(), hclust()
Imputation via Classification and Regression Trees (CART): rpart, caret, RWeka
Determination of imputation matching variables via Random Forests: randomForest
Creation of homogeneous imputation classes via CART: rpart
Derivation of edit rules via association analysis: arules

11. Record Linkage Weighting vector classification:
The last major step in record linkage or record de-duplication
could be understood as a classification problem
In R: rpart, bagging() in package ipred, ada, functions svm() and nnet() in package e1071

12. Other Methods
Questionnaire consolidation via cluster analysis: class, klaR, cluster...
Forming non-response weighting groups via classification trees: rpart, tree, caret
Non-respondent prediction via classification trees: rpart, tree, caret
Analysis of reporting errors via classification trees: rpart, tree, caret
Substitutes for surveys via internet scraping: scrapeR, rvest
Tax evader detection via k-nearest neighbours: class, kknn
Crop yield estimation via image processing on satellite imaging data: is this ML?

13. Fernandez-Delgado, Cernadas, Barro (2014)
Do we Need Hundreds of Classiers to Solve Real World Classication Problems?
Fernandez-Delgado, Cernadas, Barro (2014)
Evaluate 179 classifiers arising from 17 families on 121 data sets
By far best are random forests and SVM with Gaussian kernel
Most of the best classiffiers are implemented in R and tuned using caret
seems the best alternative to select a classier implementation

14. Top 10 ML/DM Algorithms Xindong Wu and Vipin Kumar (2009)
C4.5 – generates classifiers expressed as decision trees or ruleset form
K-Means – simple iterative method to partition a given dataset into a userspecified number of clusters, k
SVM – support vector machines
Apriori - derive association rules
EM - Expectation–Maximization algorithm
PageRank - produces a static ranking of Web pages
AdaBoost – Ensemble learning
kNN - k-nearest neighbor classification
Naive Bayes – simple classifier, applying the Bayes‘ theorem with independence assumptions between the features
CART - Classification and Regression Trees
C4.5 builds decision trees from a set of training data, using the concept of information entropy . The algorithm was developed by Ross Quinlan (1993) and is an an extension of Quinlan's earlier ID3 algorithm. The decision trees generated by C4.5 can be used for classification, and thus C4.5 is often referred to as a statistical classifier.

15. The best R packages for ML
e1071: Naive Bayes, SVM, latent class analysis
rpart: regression trees
RandomForest: RF
gbm: generalized boosting models
kernlab: SVM
caret: Classification and Regression Training
neuralnet: neural networks
CRAN Task View: Machine Learning & Statistical Learning

16. Machine learning books