This presentation would give a simple introduction to perform some basic data analysis using WEKA. Since it contains an image guiding criteria it would be very easy for the beginners.

1. A SIMPLE INTRODUCTION
TO WEKA

2. Contents
 What is WEKA?
 WEKA Explorer
 Preprocessing the data
 Classification
 Clustering
 Association Rules
 Attribute Selection
 Data Visualization

3.  Waikato Environment for Knowledge Analysis
 Developed by Department of Computer Science, University of
Waikato, New Zealand.
 Weka is also a bird found only on the
islands of New Zealand.
1 What is WEKA?

4.  A collection of machine learning algorithms
for data mining tasks.
 Download and Install WEKA
Website:http://www.cs.waikato.ac.nz/~ml/weka/ind
ex.html
 Platform independent

5. Exploratory data analysis
Experimental environment
New process model inspired
interface
Command Line Interface
WEKA GUI

6. WEKA Explorer
2

7.  Preprocessing the data
 Classification
 Clustering
 Association Rules
 Attribute Selection
 Data Visualization

8. “
Pre-Processing the data
 Data can be imported from a file in various
formats.
ARFF-Attribute-Relation File Format
CSV - Comma Separated Values
 Data can be read from a URL or from a SQL
database.
 Filters are used for pre-processing

9. @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

13. 7/1/2016University of Waikato 13

14. 7/1/2016University of Waikato 14

15. 7/1/2016University of Waikato 15

16. 7/1/2016University of Waikato 16

17. 7/1/2016University of Waikato 17

18. 7/1/2016University of Waikato 18

19. 7/1/2016University of Waikato 19

20. 7/1/2016University of Waikato 20

21. 7/1/2016University of Waikato 21

22. 7/1/2016University of Waikato 22

23. 7/1/2016University of Waikato 23

24. 7/1/2016University of Waikato 24

25. 7/1/2016University of Waikato 25

26. 7/1/2016University of Waikato 26

27. 7/1/2016University of Waikato 27

28. 7/1/2016University of Waikato 28

29. 7/1/2016University of Waikato 29

30. 7/1/2016University of Waikato 30

31. 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 perceptron,
logistic regression, Bayes’ nets, …

32. age income student credit_rating buys_computer
<=30 high no fair no
<=30 high no excellent no
31…40 high no fair yes
>40 medium no fair yes
>40 low yes fair yes
>40 low yes excellent no
31…40 low yes excellent yes
<=30 medium no fair no
<=30 low yes fair yes
>40 medium yes fair yes
<=30 medium yes excellent yes
31…40 medium no excellent yes
31…40 high yes fair yes
>40 medium no excellent no
Decision Tree Induction: Training Dataset

33. age?
overcast
student? credit rating?
<=30 >40
no yes yes
yes
31..40
fairexcellentyesno
Output: A Decision Tree for “buys_computer”

34. 7/1/2016 University of Waikato 34

35. 7/1/2016 University of Waikato 35

36. 7/1/2016 University of Waikato 36

37. 7/1/2016University of Waikato 37

38. 7/1/2016University of Waikato 38

39. 7/1/2016University of Waikato 39

40. 7/1/2016University of Waikato 40

41. 7/1/2016University of Waikato 41

42. 7/1/2016University of Waikato 42

43. 7/1/2016University of Waikato 43

44. 7/1/2016University of Waikato 44

45. 7/1/2016University of Waikato 45

46. 7/1/2016University of Waikato 46

47. 7/1/2016University of Waikato 47

48. 7/1/2016University of Waikato 48

49. 7/1/2016University of Waikato 49

50. 7/1/2016University of Waikato 50

51. 7/1/2016University of Waikato 51

52. 7/1/2016University of Waikato 52

53. 7/1/2016University of Waikato 53

54. 7/1/2016University of Waikato 54

55. 7/1/2016University of Waikato 55

56. Clustering data
 Finding groups of similar instances in a
dataset
 Implemented schemes in WEKA are:
k-Means, EM, Cobweb, X-means,
FarthestFirst

57. Finding Associations
 WEKA contains an implementation of the Apriori
algorithm for learning association rules
 Works only with discrete data
 Can identify statistical dependencies between
groups of attributes:

58. 7/1/2016University of Waikato 58

59. 7/1/2016University of Waikato 59

60. 7/1/2016University of Waikato 60

61. 7/1/2016University of Waikato 61

62. 7/1/2016University of Waikato 62

63.  Used to determine the most predictive attributes
 Consists of two parts
1.) A search method : best-first, forward selection,
random, exhaustive, genetic algorithm and etc.
2.)An evaluation method : correlation-based, wrapper,
information gain an etc.
Attribute Selection

64. 7/1/2016University of Waikato 64

65. 7/1/2016University of Waikato 65

66. 7/1/2016University of Waikato 66

67. 7/1/2016University of Waikato 67

68. 7/1/2016University of Waikato 68

69. 7/1/2016University of Waikato 69

70. 7/1/2016University of Waikato 70

71. 7/1/2016University of Waikato 71

72. Data Visualization
 WEKA can visualize single attributes (1-d) and
pairs of attributes (2-d)
 Color-coded class values
 Use of “jitter” option
 “Zoom-in” function

73. 7/1/2016University of Waikato 73

74. 7/1/2016University of Waikato 74

75. 7/1/2016University of Waikato 75

76. 7/1/2016University of Waikato 76

77. 7/1/2016University of Waikato 77

78. 7/1/2016University of Waikato 78

79. 7/1/2016University of Waikato 79

80. 7/1/2016University of Waikato 80

81. 7/1/2016University of Waikato 81

82. 7/1/2016University of Waikato 82

83. The End