Data mining -- code implementation of association analysis example (Part 2)

1. Guide pack

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

2. Read the file

my_data=pd.read_excel("D:/ Great / data mining /basket.xlsx")
df_data=my_data.iloc[:,7:].copy()
df_data.head()

 3. data display

my_data.describe()

The result is shown in Fig. ：

 4. Data processing ：

dict_data={'F':False,'T':True}
df_data['fruitveg']=df_data['fruitveg'].map(dict_data)
df_data['freshmeat']=df_data['freshmeat'].map(dict_data)
df_data['dairy']=df_data['dairy'].map(dict_data)
df_data['cannedveg']=df_data['cannedveg'].map(dict_data)
df_data['cannedmeat']=df_data['cannedmeat'].map(dict_data)
df_data['frozenmeal']=df_data['frozenmeal'].map(dict_data)
df_data['beer']=df_data['beer'].map(dict_data)
df_data['wine']=df_data['wine'].map(dict_data)
df_data['softdrink']=df_data['softdrink'].map(dict_data)
df_data['fish']=df_data['fish'].map(dict_data)
df_data['confectionery']=df_data['confectionery'].map(dict_data)

give the result as follows ：

 

5. Set support to find frequent itemsets

 frequent_itemsets = apriori(df_data,min_support=0.1,use_colnames= True)

frequent_itemsets

give the result as follows ：

 

 6. 

# Find association rules , Set the minimum confidence to 0.15

rules = association_rules(frequent_itemsets,metric = 'confidence',min_threshold = 0.15)

# Set the minimum lift

rules = rules.drop(rules[rules.lift <1.0].index)

# Set the title index and print the results

rules.rename(columns = {'antecedents':'from','consequents':'to','support':'sup','confidence':'conf'},inplace = True)

rules = rules[['from','to','sup','conf','lift']]

rules

give the result as follows ：