Day1 ARM 2.ipynb

Kernel: Python 3 (ipykernel)

Association Rule Mining (ARM) is a popular unsupervised learning technique used to discover interesting relationships between variables in large datasets. The most common application of ARM is in market basket analysis, where the goal is to find associations between items that frequently co-occur in transactions.

pip install wordcloud --trusted-host pypi.org --trusted-host files.pythonhosted.org mlxtend

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

# Sample dataset: transactions represented as a list of lists
dataset = [['milk', 'bread', 'butter'],
           ['beer', 'bread'],
           ['milk', 'diapers', 'beer', 'bread'],
           ['butter', 'diapers', 'milk', 'beer', 'bread'],
           ['butter', 'diapers', 'milk', 'beer']]

# Convert dataset into a one-hot encoded DataFrame
te = TransactionEncoder()
te_ary = te.fit(dataset).transform(dataset)
df = pd.DataFrame(te_ary, columns=te.columns_)
df

# Generate frequent itemsets with a minimum support of 0.6
frequent_itemsets = apriori(df, min_support=0.6, use_colnames=True)
print("Frequent Itemsets:\n", frequent_itemsets)

Frequent Itemsets:
     support               itemsets
0       0.8                 (beer)
1       0.8                (bread)
2       0.6               (butter)
3       0.6              (diapers)
4       0.8                 (milk)
5       0.6          (beer, bread)
6       0.6        (beer, diapers)
7       0.6           (beer, milk)
8       0.6          (bread, milk)
9       0.6         (butter, milk)
10      0.6        (milk, diapers)
11      0.6  (beer, milk, diapers)

frequent_itemsets = apriori(df, min_support=0.6, use_colnames=True)

pen_book = 3
pen=4
book=3

s_b = 6/10
s_p = 7/10
print("support for book",s_b)
print("support for pen",s_p)
print("How much prob that if one buys book will also purcahse pen?")
C_p_b = 3/6
C_b_p = 4/7
print(C_p_b)
print(C_b_p)
Lift= C_p_b/s_p 

support for book 0.6
support for pen 0.7
How much prob that if one buys book will also purcahse pen?
0.5
0.5714285714285714

# Generate association rules with a minimum confidence of 0.7
rules = association_rules(frequent_itemsets, metric="confidence", min_threshold=0.7)
print("\nAssociation Rules:\n", rules)

Association Rules:
         antecedents      consequents  antecedent support  consequent support  \
0            (beer)          (bread)                 0.8                 0.8   
1           (bread)           (beer)                 0.8                 0.8   
2            (beer)        (diapers)                 0.8                 0.6   
3         (diapers)           (beer)                 0.6                 0.8   
4            (beer)           (milk)                 0.8                 0.8   
5            (milk)           (beer)                 0.8                 0.8   
6           (bread)           (milk)                 0.8                 0.8   
7            (milk)          (bread)                 0.8                 0.8   
8          (butter)           (milk)                 0.6                 0.8   
9            (milk)         (butter)                 0.8                 0.6   
10           (milk)        (diapers)                 0.8                 0.6   
11        (diapers)           (milk)                 0.6                 0.8   
12     (beer, milk)        (diapers)                 0.6                 0.6   
13  (beer, diapers)           (milk)                 0.6                 0.8   
14  (milk, diapers)           (beer)                 0.6                 0.8   
15           (beer)  (milk, diapers)                 0.8                 0.6   
16           (milk)  (beer, diapers)                 0.8                 0.6   
17        (diapers)     (beer, milk)                 0.6                 0.6   

    support  confidence      lift  leverage  conviction  zhangs_metric  
0       0.6        0.75  0.937500     -0.04         0.8          -0.25  
1       0.6        0.75  0.937500     -0.04         0.8          -0.25  
2       0.6        0.75  1.250000      0.12         1.6           1.00  
3       0.6        1.00  1.250000      0.12         inf           0.50  
4       0.6        0.75  0.937500     -0.04         0.8          -0.25  
5       0.6        0.75  0.937500     -0.04         0.8          -0.25  
6       0.6        0.75  0.937500     -0.04         0.8          -0.25  
7       0.6        0.75  0.937500     -0.04         0.8          -0.25  
8       0.6        1.00  1.250000      0.12         inf           0.50  
9       0.6        0.75  1.250000      0.12         1.6           1.00  
10      0.6        0.75  1.250000      0.12         1.6           1.00  
11      0.6        1.00  1.250000      0.12         inf           0.50  
12      0.6        1.00  1.666667      0.24         inf           1.00  
13      0.6        1.00  1.250000      0.12         inf           0.50  
14      0.6        1.00  1.250000      0.12         inf           0.50  
15      0.6        0.75  1.250000      0.12         1.6           1.00  
16      0.6        0.75  1.250000      0.12         1.6           1.00  
17      0.6        1.00  1.666667      0.24         inf           1.00