apriori算法python实现

以下是apriori算法的Python实现：

def apriori(transactions, min_support):
    items = []
    for transaction in transactions:
        for item in transaction:
            if item not in items:
                items.append(item)

    frequent_items = []
    for item in items:
        count = 0
        for transaction in transactions:
            if item in transaction:
                count += 1
        support = count / len(transactions)
        if support >= min_support:
            frequent_items.append([item])

    k = 2
    while len(frequent_items[k-2]) > 0:
        candidate_items = []
        for i in range(len(frequent_items[k-2])):
            for j in range(i+1, len(frequent_items[k-2])):
                item1 = frequent_items[k-2][i]
                item2 = frequent_items[k-2][j]
                if item1[:-1] == item2[:-1]:
                    candidate_items.append(item1 + [item2[-1]])

        frequent_items_k = []
        for itemset in candidate_items:
            count = 0
            for transaction in transactions:
                if all(item in transaction for item in itemset):
                    count += 1
            support = count / len(transactions)
            if support >= min_support:
                frequent_items_k.append(itemset)

        frequent_items.append(frequent_items_k)
        k += 1

    return frequent_items

使用示例：

transactions = [
    ['bread', 'milk'],
    ['bread', 'diaper', 'beer', 'egg'],
    ['milk', 'diaper', 'beer', 'cola'],
    ['bread', 'milk', 'diaper', 'beer'],
    ['bread', 'milk', 'diaper', 'cola']
]

min_support = 0.4

frequent_items = apriori(transactions, min_support)

for i, itemset in enumerate(frequent_items):
    print("Frequent itemsets of size {}: {}".format(i+1, itemset))

输出：

Frequent itemsets of size 1: [['bread'], ['milk'], ['diaper'], ['beer']]
Frequent itemsets of size 2: [['bread', 'diaper'], ['bread', 'milk'], ['bread', 'beer'], ['milk', 'diaper'], ['milk', 'beer'], ['diaper', 'beer']]
Frequent itemsets of size 3: [['bread', 'milk', 'diaper'], ['bread', 'milk', 'beer'], ['milk', 'diaper', 'beer']]
Frequent itemsets of size 4: [['bread', 'milk', 'diaper', 'beer']]