Introduction

SSDP+ is an evolutionary approach for mining diverse and more informative subgroups focused on high dimensional data sets. SSDP+ is an extension of the SSDP model (https://github.com/tarcisiodpl/ssdp) to provide diversity in a way that explore the relation between subgroups order to generate a more informative set of patterns.

Link for SSDP+ paper:

https://ieeexplore.ieee.org/document/8477855

Video:

Rodando o SSDP+ (português): https://youtu.be/u175JvLD-38

Description folders:

SSDP+

Java implementation of SSDP+. SSDP+ was implemented as a Netbeans project. Thus, you have to:

download and install the Netbeans with Java (https://www.oracle.com/technetwork/pt/java/javase/downloads/jdk-netbeans-jsp-3413153-ptb.html)
download the SSDPplus of this repository and open in Neatbeans

Class SSDPplus contains a main method that is a self explanatory way how to run the algorithm:

 //*******************************************
 //Data set                    ***************
 //*******************************************
 String caminho = "C:\\Users\\Tarcisio  Lucas\\Documents\\NetBeansProjects\\SSDPplus\\pastas\\bases\\"; 
 //String nomeBase = "alon-clean50-pn-width-2.CSV";
 //String nomeBase = "ENEM2014_81_NOTA_10k.csv";
 String nomeBase = "matrixBinaria-Global-100-p.csv";
 String caminhoBase = caminho + nomeBase;

 //separator database (CSV files)
 D.SEPARADOR = ","; 
 //Seed
 Const.random = new Random(Const.SEEDS[0]); 
 //*******************************************
 //END Data set                    ***************
 //*******************************************
 

 //*******************************************
 //SSDP+ parameters            ***************
 //*******************************************
 //k: number of subgroups
 int k = 5; 
 //Evaluation metric
 String tipoAvaliacao = Avaliador.METRICA_AVALIACAO_WRACC; 
     //String tipoAvaliacao = Avaliador.METRICA_AVALIACAO_QG;
 //ks: cache size
 Pattern.maxSimulares = 2; 
 //min_similarity
 double similaridade = 0.10;
 //Similarity function
 Pattern.medidaSimilaridade = Const.SIMILARIDADE_JACCARD; //similarity function (default JACCARD)
 //Target (atributevalue)
 String target = "p";
 
 //*******************************************
 //END SSDP+ parameters            ***************
 //*******************************************
 
 //It is not about the SSDP+
 Pattern.ITENS_OPERATOR = Const.PATTERN_AND;
 
 //max time simulation in second (-1 for infinity)
 double maxTimeSecond =  -1;      
 
 System.out.println("Loading data set...");
 D.CarregarArquivo(caminhoBase, D.TIPO_CSV); //Loading data set        
 D.GerarDpDn(target);
 //"6,80,104,116,134,145,151,153,156,256"; //target value
 //D.valorAlvo = "I-III";
 //D.valorAlvo = "IV-VII";
 
 
 
 //*******************************************
 //FILTER BY ATTRIBUTE, VALUES AND ITEMS *****
 //*******************************************
 //Filter by attribute
 //String[] filtrarAtributos = {"C009"};
 String[] filtrarAtributos = null;
 //Filter by values
 String[] filtrarValores = null;
 //String[] filtrarValores = {"", "NA"};
 //Filter by items
 String[][] filtrarAtributosValores = null;
 //String[][] filtrarAtributosValores = new String[2][2];
 //filtrarAtributosValores[0][0] = "D001";
 //filtrarAtributosValores[0][1] = "2";
 //filtrarAtributosValores[1][0] = "E01002";
 //filtrarAtributosValores[1][1] = "8";
 //*******************************************
 //EDN FILTER BY ATTRIBUTE, VALUES AND ITEMS *****
 //*******************************************

experiments:

This folder contains the results of the experiments

data sets:

This folder contains part of the databases used in the experiments