Download presentation
Presentation is loading. Please wait.
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
2
6/21/2015University of Waikato2 WEKA: the bird Copyright: Martin Kramer (mkramer@wxs.nl)
3
6/21/2015University of Waikato3 WEKA: the software Machine learning/data mining software written in Java (distributed under the GNU Public License) 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
6/21/2015University of Waikato4 @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
5
6/21/2015University 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
6/21/2015University of Waikato6
7
6/21/2015University of Waikato7 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” WEKA contains filters for: Discretization, normalization, resampling, attribute selection, transforming and combining attributes, …
8
6/21/2015University of Waikato8
9
6/21/2015University of Waikato9
10
6/21/2015University of Waikato10
11
6/21/2015University of Waikato11
12
6/21/2015University of Waikato12
13
6/21/2015University of Waikato13
14
6/21/2015University of Waikato14
15
6/21/2015University of Waikato15
16
6/21/2015University of Waikato16
17
6/21/2015University of Waikato17
18
6/21/2015University of Waikato18
19
6/21/2015University of Waikato19
20
6/21/2015University of Waikato20
21
6/21/2015University of Waikato21
22
6/21/2015University of Waikato22
23
6/21/2015University of Waikato23
24
6/21/2015University of Waikato24
25
6/21/2015University of Waikato25
26
6/21/2015University of Waikato26
27
6/21/2015University of Waikato27
28
6/21/2015University of Waikato28
29
6/21/2015University of Waikato29 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, … “Meta”-classifiers include: Bagging, boosting, stacking, error-correcting output codes, locally weighted learning, …
30
6/21/2015University of Waikato30
31
6/21/2015University of Waikato31
32
6/21/2015University of Waikato32
33
6/21/2015University of Waikato33
34
6/21/2015University of Waikato34
35
6/21/2015University of Waikato35
36
6/21/2015University of Waikato36
37
6/21/2015University of Waikato37
38
6/21/2015University of Waikato38
39
6/21/2015University of Waikato39
40
6/21/2015University of Waikato40
41
6/21/2015University of Waikato41
42
6/21/2015University of Waikato42
43
6/21/2015University of Waikato43
44
6/21/2015University of Waikato44
45
6/21/2015University of Waikato45
46
6/21/2015University of Waikato46
47
6/21/2015University of Waikato47
48
6/21/2015University of Waikato48
49
6/21/2015University of Waikato49
50
6/21/2015University of Waikato50
51
6/21/2015University of Waikato51
52
6/21/2015University of Waikato52
53
6/21/2015University of Waikato53 Explorer: clustering data WEKA contains “clusterers” for finding groups of similar instances in a dataset Implemented schemes are: k-Means, EM, Cobweb, X-means, FarthestFirst Clusters can be visualized and compared to “true” clusters (if given) Evaluation based on loglikelihood if clustering scheme produces a probability distribution
54
6/21/2015University of Waikato54
55
6/21/2015University of Waikato55
56
6/21/2015University of Waikato56
57
6/21/2015University of Waikato57
58
6/21/2015University of Waikato58
59
6/21/2015University of Waikato59
60
6/21/2015University of Waikato60
61
6/21/2015University of Waikato61
62
6/21/2015University of Waikato62
63
6/21/2015University of Waikato63
64
6/21/2015University of Waikato64
65
6/21/2015University of Waikato65
66
6/21/2015University of Waikato66
67
6/21/2015University of Waikato67
68
6/21/2015University of Waikato68
69
6/21/2015University of Waikato69 Explorer: 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: milk, butter bread, eggs (with confidence 0.9 and support 2000) Apriori can compute all rules that have a given minimum support and exceed a given confidence
70
6/21/2015University of Waikato70
71
6/21/2015University of Waikato71
72
6/21/2015University of Waikato72
73
6/21/2015University of Waikato73
74
6/21/2015University of Waikato74
75
6/21/2015University of Waikato75
76
6/21/2015University of Waikato76
77
6/21/2015University of Waikato77 Conclusion: try it yourself! WEKA is available at http://www.cs.waikato.ac.nz/ml/weka Also has a list of projects based on WEKA
Similar presentations
