Download Weka tutorial - Personal Web Pages

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
For ITCS 6265/8265
Fall 2009
TA: Fei Xu
UNC Charlotte
Contents
 What is weka?
 RAFF data format
 Interface
 Explorer
 …
 Trouble shooting
WEKA: the bird
Copyright: Martin Kramer ([email protected])
Weka: Data Mining Software
 Weka is a collection of machine learning algorithms
for data mining tasks.
 The algorithms can either be applied directly to a
dataset or called from your own Java code.
 Weka contains tools for data pre-processing,
classification, regression, clustering, association rules,
and visualization.
 Weka is open source software in JAVA issued under the
GNU General Public License.
 http://www.cs.waikato.ac.nz/ml/weka/
WEKA only deals with “flat” files
 Weka has it own file format: Attribute-Relation File
Format (ARFF)
 Header section and Data section
 Supported attributes:
 numeric, nominal, string, date
 Details at:
 http://www.cs.waikato.ac.nz/~ml/weka/arff.html
Example
@RELATION auto-mpg
@ATTRIBUTE mpg REAL
@ATTRIBUTE cylinders REAL
@ATTRIBUTE displacement REAL
@ATTRIBUTE horsepower REAL
@ATTRIBUTE weight REAL
@ATTRIBUTE acceleration REAL
@ATTRIBUTE year REAL
@ATTRIBUTE origin {1,2,3} % 1 = usa; 2 = europe; 3 = japan
@DATA
15.0 8 350.0
18.0 8 318.0
17.0 8 302.0
15.0 8 429.0
14.0 8 440.0
165.0
150.0
140.0
198.0
215.0
3693.
3436.
3449.
4341.
4312.
11.5
11.0
10.5
10.0
8.5
70 1
70 1
70 1
70 1
70 1
Explorer Interface
 Explorer
 pre-processing the data
 building “classifiers” (demo)
 clustering data
 finding associations
 attribute selection
 data visualization
Other interface
Experimenter makes it easy
to compare the performance
of different learning
schemes
Java-Beans-based interface
for setting up and running
machine learning
experiments.
Command line interface
More at: http://www.cs.waikato.ac.nz/ml/weka/index_documentation.html
Trouble shooting
 OutOfMemoryException
 Find “RunWeka.ini” under weka installation directory,
default location in windows is “C:\Program Files\Weka3-6”.
 Find “maxheap” and change the value to proper size, for
example 512M
 More at: http://weka.wikispaces.com/Troubleshooting
Related documents