USML5

import pandas as pd
from mlxtend.preprocessing import TransactionEncoder
from mlxtend.frequent_patterns import fpgrowth, association_rules

# 1. Define your raw retail transaction data
dataset = [
    ['milk', 'bread', 'nut'],
    ['milk', 'bread', 'apple'],
    ['milk', 'bread', 'nut'],
    ['bread', 'apple'],
    ['milk', 'nut'],
]


te = TransactionEncoder()
te_ary = te.fit(dataset).transform(dataset)
df = pd.DataFrame(te_ary, columns=te.columns_)

print("Encoded DataFrame Structure")
print(df)

frequent_itemsets = fpgrowth(df, min_support=0.4, use_colnames=True)

print("Mined Frequent Itemsets")
print(frequent_itemsets)

rules = association_rules(frequent_itemsets, metric="confidence", min_threshold=0.7)

print("Generated Association Rules")
print(rules[['antecedents', 'consequents', 'support', 'confidence', 'lift']])