SEQUAL ====== The description of each script is as follows: * createfeatures.m Inputs: ../data/SCF (the folder where SCF files reside) ../data/labels.txt file Output: features.arff Prerequisite: - SCF files in ../data/SCF - MATLAB and MATLAB Bioinformatics Toolbox It reads the SCF files on ../data/SCF folder and their associated labels ../data/labels.txt in order to create the features. The resulting features are saved into features.arff. The script uses scfread function bundled in MATLAB Bioinformatics Toolbox. If you want to use the script to apply to your trace files, then unfortunately you need to obtain this toolbox. If you are only interested in the our results, then you don't need to run the script. Instead you can use already created features features.arff * features.arff This is the set of features created by transforming SCF files by using createfeatures.m script. It is formatted in ARFF format so that it is readable in WEKA toolbox. * weka.jar This is the bare bones JAR file of WEKA platform downloded from http://www.cs.waikato.ac.nz/~ml/weka/snapshots/developer-branch.zip I copy this file in case you have some difficulties installing WEKA. * gridSearch.jar This is the gridSearch plugin used for searching the feasible parameters of SVM RBF and Polynomial Kernal. It can be downloaded from WEKA sourceforge web site, but I included it here for you don't have to search for the appropriate file. * knowledgeflow.kfml Input: ../src/features.arff Prerequisite: - The Java Runtime Environment should be installed. - WEKA developer branch (http://www.cs.waikato.ac.nz/~ml/weka/snapshots/developer-branch.zip) or you can use the one in lib/weka.jar Output: The output is displayed in the TextViewer component in the WEKA KnowdledgeFlow Environment Running: - Download the followings to a same folder - ../src/features.arff - ../src/weka.jar - ../src/gridSearch.jar - ../src/knowledgeflow.kfml - First open WEKA GUI Chooser by using the command "java -classpath weka.jar:gridSearch.jar weka.gui.GUIChooser" (linux, Mac) "java -classpath weka.jar;gridSearch.jar weka.gui.GUIChooser" (Windows) - Click KnowledgeFlow button. - By clicking open button or press Ctrl-O - Choose knowledgeflow.kfml file - Click the run button - Wait for the run - See the results on the TextViewer