用python实现apriori算法

好的，以下是用 Python 实现 Apriori 算法的示例代码：

def load_dataset():
    """
    加载数据集。
    """
    dataset = [['milk', 'bread', 'eggs'],
               ['milk', 'bread', 'butter'],
               ['milk', 'butter'],
               ['bread', 'butter'],
               ['bread', 'eggs']]
    return dataset


def create_c1(dataset):
    """
    创建 C1 集合，即所有项的单元素集合。
    """
    c1 = set()
    for transaction in dataset:
        for item in transaction:
            c1.add(frozenset([item]))  # 使用 frozenset 可以使集合可以作为字典的键
    return c1


def scan_dataset(dataset, candidate_set, min_support):
    """
    扫描数据集，计算每个项集的支持度，并返回支持度大于等于 min_support 的项集。
    """
    support_count = {}
    for transaction in dataset:
        for candidate in candidate_set:
            if candidate.issubset(transaction):
                support_count[candidate] = support_count.get(candidate, 0) + 1

    num_items = float(len(dataset))
    frequent_set = []
    support_data = {}
    for itemset in support_count:
        support = support_count[itemset] / num_items
        if support >= min_support:
            frequent_set.append(itemset)
            support_data[itemset] = support

    return frequent_set, support_data


def apriori_gen(frequent_set, k):
    """
    由频繁项集生成候选项集。
    """
    candidate_set = []
    len_fset = len(frequent_set)
    for i in range(len_fset):
        for j in range(i + 1, len_fset):
            l1 = list(frequent_set[i])[:k - 2]
            l2 = list(frequent_set[j])[:k - 2]
            l1.sort()
            l2.sort()
            if l1 == l2:
                candidate_set.append(frequent_set[i] | frequent_set[j])

    return candidate_set


def apriori(dataset, min_support=0.5):
    """
    Apriori 算法。
    """
    c1 = create_c1(dataset)
    frequent_set, support_data = scan_dataset(dataset, c1, min_support)
    frequent_sets = [frequent_set]
    k = 2
    while len(frequent_sets[k - 2]) > 0:
        candidate_set = apriori_gen(frequent_sets[k - 2], k)
        frequent_set, support_data_k = scan_dataset(dataset, candidate_set, min_support)
        support_data.update(support_data_k)
        frequent_sets.append(frequent_set)
        k += 1

    return frequent_sets, support_data


if __name__ == '__main__':
    dataset = load_dataset()
    frequent_sets, support_data = apriori(dataset)
    print('频繁项集：')
    print(frequent_sets)
    print('支持度：')
    print(support_data)

这里的数据集是一个简单的例子，你可以根据自己的实际需求修改数据集和参数。