1. Department of Computer Science, University of Waikato, New Zealand Eibe Frank WEKA: A Machine Learning Toolkit The Explorer Classification and Regression Clustering Association Rules Attribute Selection Data Visualization The Experimenter The Knowledge Flow GUI Conclusions Machine Learning with WEKA based on notes by

2. 4/20/2014University of Waikato2 WEKA: the bird Copyright: Martin Kramer (mkramer@wxs.nl)

3. 4/20/2014University of Waikato3 WEKA: the software Machine learning/data mining software written in Java (distributed under the GNU Public License) Used for research, education, and applications Complements Data Mining by Witten & Frank Main features: Comprehensive set of data pre-processing tools, learning algorithms and evaluation methods Graphical user interfaces (incl. data visualization) Environment for comparing learning algorithms

4. 4/20/2014University of Waikato4 WEKA: versions There are several versions of WEKA: WEKA 3.0: book version compatible with description in data mining book 1 st edition WEKA 3.2: GUI version adds graphical user interfaces (earlier version is command-line only) WEKA 3.4 ++ on SoC linux and ISS windows This talk is based on snapshots of WEKA 3.3 … with some extra up-to-date snapshots Only changes are layout and some extras

5. 4/20/2014University of Waikato5 @relation heart-disease-simplified @attribute age numeric @attribute sex { female, male} @attribute chest_pain_type { typ_angina, asympt, non_anginal, atyp_angina} @attribute cholesterol numeric @attribute exercise_induced_angina { no, yes} @attribute class { present, not_present} @data 63,male,typ_angina,233,no,not_present 67,male,asympt,286,yes,present 67,male,asympt,229,yes,present 38,female,non_anginal,?,no,not_present... WEKA only deals with flat files

6. 4/20/2014University of Waikato6 @relation heart-disease-simplified @attribute age numeric @attribute sex { female, male} @attribute chest_pain_type { typ_angina, asympt, non_anginal, atyp_angina} @attribute cholesterol numeric @attribute exercise_induced_angina { no, yes} @attribute class { present, not_present} @data 63,male,typ_angina,233,no,not_present 67,male,asympt,286,yes,present 67,male,asympt,229,yes,present 38,female,non_anginal,?,no,not_present... WEKA only deals with flat files

7. 4/20/2014University of Waikato7

8. 4/20/2014University of Waikato8

9. 4/20/2014University of Waikato9

10. 4/20/2014University of Waikato10 Explorer: pre-processing the data Data can be imported from a file in various formats: ARFF, CSV, C4.5, binary Data can also be read from a URL or from an SQL database (using JDBC) Pre-processing tools in WEKA are called filters BUT it may be easier to reformat to ARFF yourself (write a program in python / java … or just use WordPad to type in the text – but make sure format is right!), this helps with data understanding

11. 4/20/2014University of Waikato11

12. 4/20/2014University of Waikato12

13. 4/20/2014University of Waikato13

14. 4/20/2014University of Waikato14

15. 4/20/2014University of Waikato15

16. 4/20/2014University of Waikato16

17. 4/20/2014University of Waikato17

18. 4/20/2014University of Waikato18 Explorer: building classifiers Classifiers in WEKA are models for predicting nominal or numeric quantities Implemented learning schemes include: Decision trees and lists, instance-based classifiers, support vector machines, multi-layer perceptrons, logistic regression, Bayes nets, … You explore by trying different classifiers, see which works best for you…

19. 4/20/2014University of Waikato19

20. 4/20/2014University of Waikato20

21. 4/20/2014University of Waikato21

22. 4/20/2014University of Waikato22

23. 4/20/2014University of Waikato23

24. 4/20/2014University of Waikato24

25. 4/20/2014University of Waikato25

26. 4/20/2014University of Waikato26

27. 4/20/2014University of Waikato27

28. 4/20/2014University of Waikato28

29. 4/20/2014University of Waikato29

30. 4/20/2014University of Waikato30

31. 4/20/2014University of Waikato31

32. 4/20/2014University of Waikato32

33. 4/20/2014University of Waikato33

34. 4/20/2014University of Waikato34

35. 4/20/2014University of Waikato35

36. 4/20/2014University of Waikato36

37. 4/20/2014University of Waikato37

38. 4/20/2014University of Waikato38

39. 4/20/2014University of Waikato39

40. 4/20/2014University of Waikato40

41. 4/20/2014University of Waikato41

42. 4/20/2014University of Waikato42

43. 4/20/2014University of Waikato43

44. 4/20/2014University of Waikato44

45. 4/20/2014University of Waikato45

46. 4/20/2014University of Waikato46

47. 4/20/2014University of Waikato47

48. WEKA from ISS PC 2009

53. @relation ukus @attribute center numeric @attribute centre numeric @attribute centerpercent numeric @attribute color numeric @attribute colour numeric @attribute colorpercent numeric @attribute english {UK,US} @data 1,32,3, 0,20,0, UK 0,25,0, 0,12,0, UK 9,27,25, 0,84,0, UK 0,19,0, 0,24,0, UK 0,16,0, 0,14,0, UK 0,16,0, 0,12,0, UK 0,21,0, 0,38,0, UK 0,25,0, 0,34,0, UK 2,26,7, 2,3,40, UK 2,32,5, 1,59,2, UK 31,0,100, 55,0,100, US 61,0,100, 26,0,100, US 24,0,100, 11,0,100, US 12,1,92, 21,4,84, US 8,0,100, 4,2,67, US 10,0,100, 8,0,100, US 19,0,100, 22,0,100, US 14,0,100, 7,0,100, US 14,0,100, 6,0,100, US 8,5,62, 24,0,100, US

65. @relation test @attribute center numeric @attribute centre numeric @attribute centerpercent numeric @attribute color numeric @attribute colour numeric @attribute colorpercent numeric @attribute english {UK,US} @data 10,5,33, 0,20,0, UK

70. 4/20/2014University of Waikato70 WEKA has more… Clustering data into groups Finding associations between attributes Visualisation - online analytical processing Experimenter to run and compare different MLs Knowledge Flow GUI 3 rd -party add-ons: sourceforge.net http://www.cs.waikato.ac.nz/ml/weka