$1\mathrm{.}3\mathrm{.}5$ Tasks

T$1)$ Find all itemsets in the dataset that occurs more than $30\%$ of the time. $($Dataframe

with $$support$$ and $$itemsets$$ columns$)$

$[]$:

T$2)$ Find all the rules in the frequent item sets $($not from task $1,$ use frequent$\_$itemsets$)$ that is true over $90\%$ of the time. $($Dataframe with $$antecedents$,$consequents$,$

$5$

$$support$,$ and $$confidence$$ columns$)$

$[]$:

$[]$:

$[]$:

$[]$:

T$3)$ Find all the $$rules$$ for class, which is our fraud$/$legitimate column. What item sets of transaction attributes $($V$1,$ V$2,$ V$3,...),$ leads to fraudulent transactions $($Class $==1).($Dataframe with $$antecedents$,$consequents$,$support$,$ and $$confidence$$ columns$)$

T$4)$ From the $$rules$$ for class, which $$rule$$ is most prevalent$/$most common in the dataset? $($Dataframe with $$antecedents$,$consequents$,$support$,$ and $$confidence$$ columns, AND a single row$)$

If you did not get T$3,$ you can use the rules dataframe for this task $($From the apriori algorithm section$)$

T$5)$ From the $$rules$$ for class, which rule has the highest consistency, that is$,$ for which itemset, is class almost always in the itemset. $($Dataframe with $$antecedents$,$consequents$,$support$,$ and $$confidence$$ columns, AND a single row$)$

If you did not get T$3,$ you can use the rules dataframe for this task $($From the apriori algorithm section$)$

$6$