ml10

//pip install mlxtend


import pandas as pd
from mlxtend.frequent_patterns import apriori, association_rules

# Sample dataset in the format of transactions
data = {'TransactionID': [1, 2, 3, 4],
        'Items': [['Milk', 'Bread', 'Butter'],
                  ['Beer', 'Diapers', 'Bread'],
                  ['Milk', 'Bread', 'Butter', 'Beer'],
                  ['Diapers', 'Milk', 'Bread']]}

df = pd.DataFrame(data)

# One hot encoding (each item is represented as a binary value in the transaction)
encoded_data = df['Items'].str.join('|').str.get_dummies()

# Convert the one-hot encoded DataFrame to boolean type to avoid the warning
encoded_data = encoded_data.astype(bool)

# Apply Apriori algorithm with minimum support of 50%
frequent_itemsets = apriori(encoded_data, min_support=0.5, use_colnames=True)

# Display frequent itemsets
print("Frequent Itemsets:")
print(frequent_itemsets)

# Generate association rules with minimum confidence of 70%
rules = association_rules(frequent_itemsets, metric="confidence", min_threshold=0.7)

# Display the association rules
print("\nAssociation Rules:")
print(rules)

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@


import pandas as pd
from mlxtend.frequent_patterns import apriori, association_rules

# Load your dataset
# Assuming a CSV file with 'TransactionID' and 'Items' columns where 'Items' is a list of items in each transaction
df = pd.read_csv('market_basket.csv')  # Replace with your file path

# Preview the dataset
print("First few rows of the dataset:\n", df.head())

# Perform One-Hot Encoding to convert transactions into a binary matrix
# Assume 'Items' column contains a list of items purchased in each transaction
encoded_data = df['Items'].str.join('|').str.get_dummies()

# Convert the one-hot encoded DataFrame to boolean type
encoded_data = encoded_data.astype(bool)

# Apply the Apriori algorithm with a minimum support threshold
frequent_itemsets = apriori(encoded_data, min_support=0.05, use_colnames=True)  # Adjust min_support as needed

# Display frequent itemsets
print("\nFrequent Itemsets:\n", frequent_itemsets)

# Generate association rules with a confidence threshold
rules = association_rules(frequent_itemsets, metric="confidence", min_threshold=0.6)  # Adjust min_threshold as needed

# Display association rules
print("\nAssociation Rules:\n", rules)

# Optional: Save the frequent itemsets and rules to CSV files for further analysis
frequent_itemsets.to_csv('frequent_itemsets.csv', index=False)
rules.to_csv('association_rules.csv', index=False)

# Print summary
print("\nFrequent itemsets and association rules have been saved to CSV files.")