Data Mining with Rattle and R

Dr Graham Williams is Senior Director of Analytics with the Australian Taxation Office, and previously Principal Computer Scientist for Data Mining with CSIRO. He is also Visiting Professor and Senior International Scientist with the Shenzhen Institutes of Advanced Analytics of the Chinese Academy of Sciences, Adjunct Professor, Data Mining, Fraud Prevention, Security, University of Canberra, and Adjunct Professor, Australian National University. Graham regularly teaches data mining courses and is author of the freely available, open source data mining system, Rattle. He has been involved in many data mining projects for clients from government and industry over his long career. His research developments included ensemble learning (1980's) and hot spots discovery (1990's). He is actively involved in the international artificial intelligence and data mining research communities, particularly as chair of the Pacific Asia Knowledge Discovery and Data Mining conference series and founder and co-chair of the Australasian Data Mining conference series. Graham has editted a number of books and authored many academic and industry papers and reports. His current focus is on making data mining technology readily accessible, ensuring research, innovation and discovery are repeatable and available, and encouraging the free and open sharing of knowledge.

From the book reviews: "The text does a great job of showing how to do each step using the data mining tool Rattle and related R concepts as appropriate. This makes it a great tool for someone who does not know much about R and wants to learn more about the powerful options available in R for data mining." (Roger M. Sauter, Technometrics, Vol. 54 (3), August, 2012) "This text is a manual for the impressive Rattle graphical user interface (GUI) for R, describing both the use of the GUI and the R code that is invoked to carry out the computations. ... Data analysts ... are likely to find Rattle a helpful tool that will allow them to quickly become productive with R. ... There is extensive useful practical advice on data preparation and data manipulation. ... is well suited for use in intermediate level courses on regression or classification." (John H. Maindonald, International Statistical Review, Vol. 80 (1), 2012)

• Introduction.- Getting Started.- Working with Data.- Loading Data.- Exploring Data.- Interactive Graphics.- Transforming Data.- Descriptive and Predictive Analytics.- Cluster Analysis.- Association Analysis.- Decision Trees.- Random Forests.- Boosting.- Support Vector Machines.- Model Performance Evaluation.- Deployment.