Presentation is loading. Please wait.

Presentation is loading. Please wait.

WEKA: A Practical Machine Learning Tool WEKA : A Practical Machine Learning Tool.

Similar presentations


Presentation on theme: "WEKA: A Practical Machine Learning Tool WEKA : A Practical Machine Learning Tool."— Presentation transcript:

1 WEKA: A Practical Machine Learning Tool WEKA : A Practical Machine Learning Tool

2 WEKA: A Practical Machine Learning Tool Contents  1.Introduction to Weka  2.Explorer  3.Other three main tools  4.Conclusions  5.Reference

3 WEKA: A Practical Machine Learning Tool Introduction – What is Weka?  In nature: A flightless bird with an inquisitive nature found only on the islands of New Zealand.  Actually: A practical machine learning tool developed by the University of Waikato in New Zealand. It is short for Waikato Environment for Knowledge Analysis.  Definition: A collection of machine learning algorithms for data mining tasks.  Language: It is written in Java and runs on almost any platform.  Usage: The algorithms can either be applied: (1) directly to a dataset (without writing any codes); (2) called from your own Java code.

4 WEKA: A Practical Machine Learning Tool Introduction – Weka consists of  Explorer  Experimenter  Knowledge flow  Simple Command Line Interface(CLI)  Other tools and Visualization  Java interface

5 WEKA: A Practical Machine Learning Tool Explorer  WEKA’s main graphical user interface  Gives access to all its facilities using menu selection and form filling.(Data-Preprocess/Classify/Cluster/Associate/Select Attributes/Visualize) 1.Data 2. Operations of Explorer with a Classification example.

6 WEKA: A Practical Machine Learning Tool Explorer – Data(1)  From files: CSV, ARFF, C4.5… ( no *.xls )  Data loaded from URL or DB *.xls *.csv Attribute-Class Attribute Instance Instances Tips : weather.arff ( C:/Program Files/Weka/data/ )

7 WEKA: A Practical Machine Learning Tool Explorer – Data(2) ARFF(Attribute-Relation File Format)  @relation  @attribute ① numeric (real or integer numbers) ② ③ string ④ date [ ]  @data  % notes More details: http://www.cs.waikato.ac.nz/ ~ml/weka/arff.html

8 WEKA: A Practical Machine Learning Tool Explorer – Operations with an example Input data  Data preprocess  Choose classifier  Test options  Run  Result analysis

9 WEKA: A Practical Machine Learning Tool Explorer Input data Summary Statistics Select an attribute Visualization

10 WEKA: A Practical Machine Learning Tool Explorer Tune Parameters Select a Filter Weka Filter Apply the Filter

11 WEKA: A Practical Machine Learning Tool Explorer Tune Parameters Select a Classifier Decide how to evaluate Model list Results

12 WEKA: A Practical Machine Learning Tool Right-click on model to get Menu (save, visualize, etc)

13 WEKA: A Practical Machine Learning Tool

14 Others – Experimenter  Comparing different learning algorithms  ------on different datasets  ------with various parameter settings  ------and analyzing the performance statistics Click it for Experimenter

15 WEKA: A Practical Machine Learning Tool Others – KnowledgeFlow  The KnowledgeFlow provides an alternative to the Explorer as a graphical front end to Weka's core algorithms.  The KnowledgeFlow is a work in progress so some of the functionality from the Explorer is not yet available. Click it for KnowledgyFlow

16 WEKA: A Practical Machine Learning Tool Others – Simple command line interface  All implementations of the algorithms have a uniform command- line interface.  java weka.classifiers.trees.J48 -t weather.arff Click it for Simple CLI

17 WEKA: A Practical Machine Learning Tool Conclusions 1.Explorer: Input data  Data preprocess  Choose classifier  Test options  Run  Result analysis 2.Experimenter: It is necessary for further studies. 3.Make full use of :  1. Java tips;  2. WekaManual.pdf; (C:/Program Files/Weka/ )  3. Play it yourself!

18 WEKA: A Practical Machine Learning Tool Reference  Mitchell, T. Machine Learning, 1997 McGraw Hill.  Ian H. Witten, Eibe Frank, Len Trigg, Mark Hall, Geoffrey Holmes, and Sally Jo Cunningham (1999). Weka: Practical machine learning tools and techniques with Java implementations.  Ian H. Witten, Eibe Frank (2005). Data Mining: Practical Machine Learning Tools and Techniques (Second Edition, 2005). San Francisco: Morgan Kaufmann  Weka Homepage: http://www.cs.waikato.ac.nz/~ml/weka/http://www.cs.waikato.ac.nz/~ml/weka/  Wekawiki: http://weka.wikispaces.com/http://weka.wikispaces.com/  Weka on SourceForge.net: http://sourceforge.net/projects/wekahttp://sourceforge.net/projects/weka  WekaManual.pdf (C:\Program Files\Weka-3-6\WekaManual.pdf)


Download ppt "WEKA: A Practical Machine Learning Tool WEKA : A Practical Machine Learning Tool."

Similar presentations


Ads by Google