apriori算法python实现

import numpy as np

def load_data(file_path):

    data = []

    with open(file_path, 'r') as f:

        for line in f.readlines():

            line = line.strip().split(',')

            data.append(line)

    return data

def create_C1(data):

    C1 = set()

    for transaction in data:

        for item in transaction:

            C1.add(frozenset([item]))

    return C1

def is_apriori(Ck_item, Lksub1):

    for item in Ck_item:

        sub_Ck = Ck_item - frozenset([item])

        if sub_Ck not in Lksub1:

            return False

    return True

def create_Ck(Lksub1, k):

    Ck = set()

    len_Lksub1 = len(Lksub1)

    list_Lksub1 = list(Lksub1)

    for i in range(len_Lksub1):

        for j in range(1, len_Lksub1):

            l1 = list(list_Lksub1[i])

            l2 = list(list_Lksub1[j])

            l1.sort()

            l2.sort()

            if l1[0:k-2] == l2[0:k-2]:

                Ck_item = list_Lksub1[i] | list_Lksub1[j]

                if is_apriori(Ck_item, Lksub1):

                    Ck.add(Ck_item)

    return Ck

def generate_Lk_by_Ck(data, Ck, min_support):

    Lk = set()

    len_data = len(data)

    item_count = {}

    for transaction in data:

        for item in Ck:

            if item.issubset(transaction):

                if item not in item_count:

                    item_count[item] = 1

                else:

                    item_count[item] += 1

    support_data = {key: value / len_data for key, value in item_count.items() if value / len_data >= min_support}

    for key in support_data:

        Lk.add(key)

    return Lk

def apriori(data, min_support=0.5):

    C1 = create_C1(data)

    D = list(map(set, data))

    L1, support_data = generate_Lk_by_Ck(D, C1, min_support)

    Lksub1 = L1.copy()

    L = [Lksub1]

    i = 2

    while True:

        Ci = create_Ck(Lksub1, i)

        Li, supK = generate_Lk_by_Ck(D, Ci, min_support)

        if not Li:

            break

        Lksub1 = Li.copy()

        L.append(Lksub1)

        i += 1

    return L, support_data

if __name__ == '__main__':

    file_path = 'your_file_path.csv'  # 请替换为你的数据文件路径

    data = load_data(file_path)

    L, support_data = apriori(data)

    print("频繁项集：", L)

    print("支持度数据：", support_data)