Java_DataMing.zip_Java 决策树_决策树_决策树 java

package CQU.EDU.yu.cn; import java.rmi.server.LogStream; import java.util.ArrayList; public class Deal_with_result { private ArrayList<Record> total_record_table=null; public ArrayList<Record> construct_decision_table=null; public ArrayList<Record> record_need_for_decision=null; private ArrayList entropy_array=new ArrayList (); private ArrayList temp_entropy_array=new ArrayList (); private ArrayList track_tag=new ArrayList(); private ArrayList decision_table_array=new ArrayList(); private int classify_column_number=4;//need for change in the special application; private int yes_count=0; private int no_count=0; private int sum_count=0; private double entropy=0; private int count_of_column=4; private int construct_decision_tree_dataset_row=0; private ArrayList tag_array=new ArrayList(); public static String dataset_infor[][]={ {"Sunny","Hot","High","Weak","No"}, {"Sunny","Hot","High","Strong","No"}, {"Overcast","Hot","High","Weak","Yes"}, {"Rain","Mild","High","Weak","Yes"}, {"Rain","Cool","Normal","Weak","Yes"}, {"Rain","Cool","Normal","Strong","No"}, {"Overcast","Cool","Normal","Strong","Yes"}, {"Sunny","Mild","High","Weak","No"}, {"Sunny","Cool","Normal","Weak","Yes"}, {"Rain","Mild","Normal","Weak","Yes"}, {"Sunny","Mild","Normal","Strong","Yes"}, {"Overcast","Mild","High","Strong","Yes"}, {"Overcast","Hot","Normal","Weak","Yes"}, {"Rain","Mild","High","Strong","No"}, {"Rain","Cool","High","Weak","null"}, {"Sunny","Mild","High","Strong","null"}, {"Sunny","Hot","Normal","Strong","null"}, {"Sunny","Cool","Normal","Strong","null"}, {"Rain","Cool","High","Strong","null"}, {"Rain","Mild","Normal","Strong","null"}, }; public Deal_with_result(){ total_record_table=new ArrayList<Record>(); construct_decision_table=new ArrayList<Record>(); record_need_for_decision=new ArrayList<Record>(); int column_length=dataset_infor[0].length; for(int row=0;row<dataset_infor.length;++row){ Record temp_record=new Record(); for(int column=0;column<column_length;++column){ temp_record.record.add(dataset_infor[row][column]); } total_record_table.add(temp_record); } } public ArrayList<Record> Get_construct_decision_table(){ return this.construct_decision_table; } public ArrayList Get_decision_table_array(){ return this.decision_table_array; } public ArrayList<Record> Get_record_need_for_decision(){ return this.record_need_for_decision; } public int Get_construct_decision_tree_dataset_row(){ return this.construct_decision_tree_dataset_row; } public void Distinct_decision_record(){ for(int i=0;i<total_record_table.size();++i){ // construct decision table if(total_record_table.get(i).record.get(classify_column_number).toString().equals("null")){ record_need_for_decision.add(total_record_table.get(i)); } else{ construct_decision_table.add(total_record_table.get(i)); } } construct_decision_tree_dataset_row=construct_decision_table.size(); } public void Distinct_tag_classify(ArrayList<Record> varial_array){ this.sum_count=varial_array.size(); for(int i=0;i<this.sum_count;++i){ if(varial_array.get(i).record.get(classify_column_number).toString().equals("Yes")){ ++yes_count; } else{ ++no_count; } } } public void Logic_deal_with(ArrayList<Record> varial_array,int sub_tag){ int temp_sum_count= varial_array.size(); ArrayList<Record> element_same_array=new ArrayList<Record>(); while(varial_array.size()!=0){//1 String temp_element= varial_array.get(0).record.get(sub_tag).toString(); for(int i=0;i<varial_array.size();++i){//2 if(temp_element.equals(varial_array.get(i).record.get(sub_tag).toString())){ element_same_array.add(varial_array.get(i)); varial_array.remove(i); --i; } }//2 this.yes_count=0; this.no_count=0; this.sum_count=0; Distinct_tag_classify(element_same_array); double x=(double)this.yes_count/this.sum_count; double y=(double)this.no_count/this.sum_count; if(x==0||y==0){ this.entropy=0; } else{ double percent=(double)this.sum_count/temp_sum_count; this.entropy=percent*(-(x)*(Math.log10(x)/Math.log10(2))-(y)*(Math.log10(y)/Math.log10(2))); } temp_entropy_array.add(this.entropy); element_same_array.removeAll( element_same_array ); }//1 //Get the Entropy of one Column try{ double column_entropy=Double.parseDouble( temp_entropy_array.get(0).toString()); for(int i=1;i< temp_entropy_array.size();++i){ column_entropy-=Double.parseDouble( temp_entropy_array.get(i).toString()); } this.entropy_array.add(sub_tag); this.entropy_array.add(column_entropy); this.temp_entropy_array.removeAll(temp_entropy_array); } catch(Exception e){ System.out.println("TTT"); System.exit(1); } } public void Product_decision_table(ArrayList<Record> varial_array){ //求取最重要的列属性 for(int i=0;i<this.count_of_column;++i){//1 boolean is_different=true; for(int j=0;is_different&&j<this.tag_array.size();++j){//2 if(i==Integer.parseInt(this.tag_array.get(j).toString())){ is_different=false; } }//2 // if(is_different){ this.yes_count=0; this.no_count=0; this.sum_count=0; Distinct_tag_classify( varial_array); double x=(double)this.yes_count/this.sum_count; double y=(double)this.no_count/this.sum_count; if(x==0||y==0){ this.temp_entropy_array.add(0); } else{ this.entropy=(-(x)*(Math.log10(x)/Math.log10(2))-(y)*(Math.log10(y)/Math.log10(2))); this.temp_entropy_array.add(this.entropy); } ArrayList<Record> destinate_array=new ArrayList<Record>(); Copy_array(varial_array, destinate_array); Logic_deal_with( destinate_array, i); } }//1 //Get the Maxmum Entropy in all columns double temp_value=0.0;//Double.parseDouble(this.entropy_array.get(1).toString()); int temp_index=0; for(int i=1;i<this.entropy_array.size();i+=2){ if(temp_value<=Double.parseDouble(this.entropy_array.get(i).toString())){ temp_value=Double.parseDouble(this.entropy_array.get(i).toString()); temp_index=i; } } int real_important_index=Integer.parseInt(this.entropy_array.get(temp_index-1).toString()); this.entropy_array.removeAll(this.entropy_array); this.decision_table_array.add(real_important_index); this.track_tag.add(real_important_index); /* * */ Important_column_deal_with(varial_array,real_important_index); } //copy array public void Copy_array(ArrayList<Record>source_array,ArrayList<Record>destinate_array){ for(int i=0;i<source_array.size();++i){ destinate_array.add(source_array.get(i)); } } public void Important_column_deal_with(ArrayList<Record> varial_array,int sub_tag){ int temp_sum_count= varial_array.size(); ArrayList<Record> element_same_array=new ArrayList<Record>(); while(varial_array.size()!=0){//1 String temp_element= varial_array.get(0).record.get(sub_tag).toString(); for(int i=0;i<varial_array.size();++i){//2 if(temp_element.equals(varial_array.get(i).record.get(sub_tag).toString())){ element_same_array.add(varial_array.get(i)); varial_array.remove(i); --i; } }//2 this.yes_count=0; this.no_count=0; this.sum_count=0; Distinct_tag_classify(element_same_array); double x=(double)this.yes_count/this.sum_count; double y=(double)this.no_count/this.sum_count; if(x==0||y==0){ this.entropy=0; } else{ this.entropy=(-(x)*(Math.log10(x)/Math.log10(2))-(y)*(Math.log10(y)/Math.log10(2))); } if(this.entropy==0){ this.decision_table_array.add(temp_element); temp_element= element_same_array.get(0).record.get(this.classify_column_number).toString(); this.decision_table_array.add(temp_element); element_same_array.removeAll(element_same_array); } else{ this.decision_table_ar