Question

我遇到了两个哈希集相交的问题。我不知道为什么一直是0，因此，相似度值为0。

无论如何，我发布了包含交叉点的方法。对应工会的部分工作正常。我很感激你的帮助。

public <T> double Similarity (Set<T> s1, Set<T> s2){

    if (s1.isEmpty() || s2.isEmpty()){
        return 0.0;
    }
    double similarity;
    int unionValue;
    int intersectionValue;

    intersectionValue=intersection(s1, s2);
    unionValue=union(s1, s2);       
    similarity=(double)intersectionValue/(double)unionValue;

    System.out.println(unionValue+"    "+intersectionValue);
    System.out.println(similarity);

    return similarity;
}



public <T> int intersection (Set<T> s1, Set<T> s2){

    Set<T> intersection = new HashSet<T>(s1);
    intersection.retainAll(s2);
    return intersection.size();
}

public <T> int union (Set<T> s1, Set<T> s2){

    Set<T> union = new HashSet<T>(s1);
    union.addAll(s1);

    for (T apiSwRepo : s2){
        if (union.toString().contains(apiSwRepo.toString())){
        }
        else
            union.add(apiSwRepo);
    }
    return union.size();

}

Answer 1

你的问题在你的比较中。

if (union.toString().contains(apiSwRepo.toString())) {
} else
    union.add(apiSwRepo);

在上面的代码中，您将Set转换为String，然后调用contains（）。你应该在union本身上使用.contains（），如下所示：

if (union.contains(apiSwRepo)) {
} else
    union.add(apiSwRepo);

我还想建议如下简化：

union.retainAll(s2);

因为你有两个集合，你将从那个简单的陈述中得到联合。

Answer 2

I don't know the reason yet. But I solved it making this changes to intersection method:

public <T> int intersection (Set<T> s1,Set<T> s2){

    Set<T> intersection = new HashSet<T>();
    for (T apiSwRepo:s2){
        if (s1.toString().contains(apiSwRepo.toString())){
            intersection.add(apiSwRepo);
        }           
    }
  return intersection.size();
}

As you can see, I did not use the intersection.retainAll because with my data it did not work, but with Set<String> test, it did.

Just to be clear, my data-set was API information of a HashMap

HashMap<Integer, Set<API>> apisPerSystem = new HashMap<Integer, Set<API>>()

Java中两个集合的交集错误

2 个答案: