人工智能和机器学习之关联规则学习算法：H-Mine算法的优化策略.docx

['milk', 'bread', 'eggs'],

['milk', 'bread'],

['bread', 'eggs'],

['milk', 'eggs'],

['milk', 'bread', 'eggs'],

['bread', 'eggs'],

['milk', 'bread', 'eggs'],

['milk', 'bread'],

['bread', 'eggs'],

['milk', 'eggs']

for transaction in data:

for itemset in itertools.combinations(transaction, 2):

hash_tree[itemset] += 1

return {k: v for k, v in hash_tree.items() if v >= min_support}

def mine_frequent_itemsets(data, min_support):

hash_tree = build_hash_tree(data, min_support)

frequent_itemsets = []

for itemset, support in hash_tree.items():

frequent_itemsets.append((itemset, support))

print(f"{itemset}: {support}")

min_confidence = 0.7

confidence = support / sum([v for k, v in hash_tree.items() if set(k).issubset(set(anteced

ent))])

if confidence >= min_confidence:

rules.append(((antecedent, consequent), confidence))

return rules

association_rules = generate_association_rules(frequent_itemsets, min_confidence)

for rule, confidence in association_rules:

print(f"{rule[0]} -> {rule[1]}: {confidence}")