Transcript Document
WEKA Mark Hall Department of Computer Science, University of Waikato, New Zealand WEKA - what is it? History WEKA’s GUI, old and new Impact Downloads Projects based on WEKA The project today Plans for the future Weka: the bird Copyright: Martin Kramer ([email protected]) 7/17/2015 University of Waikato 2 7/17/2015 University of Waikato 3 7/17/2015 University of Waikato 4 7/17/2015 University of Waikato 5 Hamilton 7/17/2015 University of Waikato 6 WEKA: the software Waikato Environment for Knowledge Analysis Collection of state-of-the-art machine learning algorithms and data processing tools implemented in Java Support for the whole process of experimental data mining Released under the GPL Preparation of input data Statistical evaluation of learning schemes Visualization of input data and the result of learning Used for education, research and applications Complements “Data Mining” by Witten & Frank 7/17/2015 University of Waikato 7 Main Features 49 data preprocessing tools 76 classification/regression algorithms 8 clustering algorithms 15 attribute/subset evaluators + 10 search algorithms for feature selection 3 algorithms for finding association rules 3 graphical user interfaces 7/17/2015 “The Explorer” (exploratory data analysis) “The Experimenter” (experimental environment) “The KnowledgeFlow” (new process model inspired interface) University of Waikato 8 History Project funded by the NZ government since 1993 7/17/2015 Develop state-of-the art workbench of data mining tools Explore fielded applications Develop new fundamental methods University of Waikato 9 History - timeline Late 1992 - funding was applied for by Ian Witten 1993 - development of the interface and infrastructure WEKA acronym coined by Geoff Holmes WEKA’s file format “ARFF” was created by Andrew Donkin ARFF was rumored to stand for Andrew’s Ridiculous File Format Sometime in 1994 - first internal release of WEKA TCL/TK user interface + learning algorithms written mostly in C Very much beta software Changes for the b1 release included (among others): “Ambiguous and Unsupported menu commands removed.” “Crashing processes handled (in most cases :-)” October 1996 - first public release of WEKA (v 2.1) 7/17/2015 University of Waikato 10 History - timeline July 1997 - WEKA 2.2 Early 1997 - decision was made to rewrite WEKA in Java Originated from code written by Eibe Frank for his PhD Originally codenamed JAWS (JAva Weka System) May 1998 - WEKA 2.3 Schemes: 1R, T2, K*, M5, M5Class, IB1-4, FOIL, PEBLS, support for C5 Included a facility (based on unix makefiles) for configuring and running large scale experiments Last release of the TCL/TK-based system Mid 1999 - WEKA 3 (100% Java) released 7/17/2015 Version to complement the Data Mining book Development version (including GUI) University of Waikato 11 Back then… 7/17/2015 University of Waikato 12 Today: 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 databases using JDBC Pre-processing tools in WEKA are called “filters” WEKA contains filters for: 7/17/2015 Discretization, normalization, resampling, attribute selection, attribute combination, … University of Waikato 15 Explorer: Building classification models “Classifiers” in WEKA are models for predicting nominal or numeric quantities Implemented schemes include: Decision trees and lists, instance-based classifiers, support vector machines, multi-layer perceptrons, logistic regression, Bayes’ nets, … “Meta”-classifiers include: 7/17/2015 Bagging, boosting, stacking, error-correcting output codes, data cleansing, … University of Waikato 16 Explorer: classification 7/17/2015 University of Waikato 17 Explorer: classification 7/17/2015 University of Waikato 18 Explorer: classification 7/17/2015 University of Waikato 19 Explorer: classification 7/17/2015 University of Waikato 20 Explorer: classification 7/17/2015 University of Waikato 21 Explorer: classification 7/17/2015 University of Waikato 22 Explorer: classification 7/17/2015 University of Waikato 23 Explorer: classification 7/17/2015 University of Waikato 24 Explorer: classification 7/17/2015 University of Waikato 25 KnowledgeFlow: process flows 7/17/2015 University of Waikato 26 KnowledgeFlow: batch processing 7/17/2015 University of Waikato 27 KnowledgeFlow: batch processing 7/17/2015 University of Waikato 28 KnowledgeFlow: incremental processing 7/17/2015 University of Waikato 29 Experimenter 7/17/2015 University of Waikato 30 Experimenter 7/17/2015 University of Waikato 31 Experimenter 7/17/2015 University of Waikato 32 Impact - downloads 7/17/2015 University of Waikato 33 Projects based on WEKA Incorporate/wrap WEKA GRB Tool Shed - a tool to aid gamma ray burst research YALE - facility for large scale ML experiments GATE - NLP workbench with a WEKA interface Judge - document clustering and classification Extend/modify WEKA BioWeka - extension library for knowledge discovery in biology WekaMetal - meta learning extension to WEKA Weka-Parallel - parallel processing for WEKA Grid Weka - grid computing using WEKA Weka-CG - computational genetics tool library 7/17/2015 University of Waikato 34 The WEKA Project Today FRST funding for the next two years Goal of the project remains the same People 6 staff 2 postdocs 3 PhD students 3 MSc students 2 research programmers 7/17/2015 University of Waikato 35 The Future Continue to develop and support WEKA MOA (Massive Online Analysis) Framework that supports learning from data streams The Moa (another native NZ bird) is not only flightless, like the Weka, but also extinct First public release, probably this Christmas, or perhaps Thanksgiving (as it’s just another turkey) MILK Facilities for data generation, experimental analysis, learning algorithms, etc. Multi-Instance Learning Kit Proper 7/17/2015 Propositionalization toolbox for WEKA University of Waikato 36