Apriori算法求频繁项目集的java实现

伪代码

算法：Apriori
输入：D - 事务数据库；min_sup - 最小支持度计数阈值
输出：L - D中的频繁项集
方法：
     L1=find_frequent_1-itemsets(D); // 找出所有频繁1项集
     For(k=2;Lk-1!=null;k++){
        Ck=apriori_gen(Lk-1); // 产生候选，并剪枝
        For each 事务t in D{ // 扫描D进行候选计数
            Ct =subset(Ck,t); // 得到t的子集
            For each 候选c 属于 Ct
                         c.count++;
        }
        Lk={c属于Ck | c.count>=min_sup}
}
Return L=所有的频繁集；

Procedure apriori_gen(Lk-1:frequent(k-1)-itemsets)
      For each项集l1属于Lk-1
              For each项集 l2属于Lk-1
                       If((l1[1]=l2[1])&&( l1[2]=l2[2])&&…….
&& (l1[k-2]=l2[k-2])&&(l1[k-1]<l2[k-1])) then{
                   c=l1连接l2 //连接步：产生候选
                   if has_infrequent_subset(c,Lk-1) then
                       delete c; //剪枝步：删除非频繁候选
                   else add c to Ck;
                  }
          Return Ck;

Procedure has_infrequent_sub(c:candidate k-itemset; Lk-1:frequent(k-1)-itemsets)
        For each(k-1)-subset s of c
            If s不属于Lk-1 then
               Return true;
        Return false;

Java实现

import java.util.*;

public class dmm {
    int zcd = 2; //最小支持度
    double kxd = 0.7; //
    String fg = ";";  //每个事务的分隔符
    Map<String, Integer> frenquentResultMap=new TreeMap<>();   //频繁项目集的结果集，按字典序排序
    public  ArrayList<String> findFrenquentOneSet(ArrayList<String> date) {
        Map<String, Integer> map = new HashMap<>();
        ArrayList<String> result=new ArrayList<>();
        for (String s : date) {                                       //获得C1
            String[] ss = s.split(fg);
            for (String c : ss) {
                if (!(map.containsKey(c)))
                    map.put(c, 1);
                else
                    map.put(c, map.get(c) + 1);
            }
        }

        for (String p : map.keySet()) {                                //获得L1
            if (map.get(p) >= zcd) {
                frenquentResultMap.put(p, map.get(p));
                result.add(p+fg);
            }
        }
        return result;                                                  //返回L1
    }

    public ArrayList<String> findFrenquentNSet(ArrayList<String> Nset, ArrayList<String> date) {    //传入数据集和CN
        ArrayList<String> result=new ArrayList<>();
        for (String s : Nset) {
            String[] s1 = s.split(fg);
            for (String ss : Nset) {
                String zuhe = "";
                String[] s2 = ss.split(fg);
                int n=0;
                for(int i=0;i< s2.length;i++){
                    if ((i!= s2.length-1&&s1[i].compareTo(s2[i])==0)||(i== s2.length-1&&s1[i].compareTo(s2[i])<0)) {
                        zuhe += s1[i] + fg;
                        n++;
                    }
                    else
                        break;;
                }
                if(n==s2.length){
                    zuhe+=s2[s2.length-1]+fg;
                    result.add(zuhe);
                }
            }
        }
        return result;                              //返回LN+1
    }
    public ArrayList<String> countFrenquentmembers(ArrayList<String> date, ArrayList<String> Nset) {
        ArrayList<String> result=new ArrayList<>();
        for (String s : Nset) {
            int count = 0;
            String[] s1 = s.split(fg);
            for (String ss : date) {
                String[] s2 = ss.split(fg);
                int n = 0;
                for (String c : s1) {
                    for (String cc : s2) {
                        if (c .compareTo(cc)==0)
                            n++;
                    }
                }
                if (n !=s1.length)
                    continue;
                count++;
            }
            if (count >=zcd) {
                frenquentResultMap.put(s, count);
                result.add(s);
            }
        }
        return result;                                  //返回CN
    }

    public static void main(String[] args) {
        ArrayList<String> dataList = new ArrayList<>();
        dataList.add("1;2;5;");
        dataList.add("2;4;");
        dataList.add("2;3;");
        dataList.add("1;2;4;");
        dataList.add("1;3;");
        dataList.add("2;3;");
        dataList.add("1;3;");
        dataList.add("1;2;3;5;");
        dataList.add("1;2;3;");
        System.out.println("====数据集合==========");
        for(String string:dataList)
        {
            System.out.println(string);
        }
        dmm aprior=new dmm();
        ArrayList<String> arr=aprior.findFrenquentOneSet(dataList);
        while(aprior.findFrenquentNSet(arr,dataList).size()!=0){
            arr=aprior.countFrenquentmembers(dataList,aprior.findFrenquentNSet(arr,dataList));
        }
        System.out.println("====频繁项集==========");
        for(String key:aprior.frenquentResultMap.keySet())
        {
            System.out.println(key+" : "+aprior.frenquentResultMap.get(key));
        }
    }
}

结果展示