数仓题2

最小支持度:当前集合在整个集合中出现的概率
最小置信度:A→B,即P(B|A),在A发生的条件下发生B的概率
C表示候选项集、L表示频繁项集

最小支持度>=30%表示10条数据中至少出现3次
C1:

ItemSet	support
a	5
b	7
c	5
d	9
e	6

L1:

ItemSet	support
a	5
b	7
c	5
d	9
e	6

C2:

ItemSet
{a,b}
{a,c}
{a,d}
{a,e}
{b,c}
{b,d}
{b,e}
{c,d}
{c,e}
{d,e}

L2:

ItemSet	support
{a,b}	3
{a,d}	4
{a,e}	4
{b,c}	3
{b,d}	6
{b,e}	4
{c,d}	4
{d,e}	6

C3:

ItemSet
{a,b,c}
{a,b,d}
{a,b,e}
{a,d,e}
{b,c,d}
{b,c,e}
{b,d,e}
{c,d,e}

L3:

ItemSet	support
{a,d,e}	4
{b,d,e}	4

C4:

ItemSet
{a,b,d,e}

L4:为空

{a,b}非空子集:a,b a->b:60% b->a:3/7
{a,d}非空子集:a,d a->d:80% d->a:4/9
{a,e}非空子集:a,e a->e:80% e->a:2/3
{b,c}非空子集:b,c b->c:3/7 c->b:60%
{b,d}非空子集:b,d b->d:6/7 d->b:2/3
{b,e}非空子集:b,e b->e:4/7 e->b:2/3
{c,d}非空子集:c,d c->d:80% d->c:4/9
{d,e}非空子集:d,e d->e:2/3 e->d:100%

{a,d,e}非空子集:ad,ae,de,a,d,e
a->de:80% de->a:2/3
d->ae:4/9 ae->d:100%
e->ad:2/3 ad->e:100%

{b,d,e}非空子集:bd,be,de,b,d,e
b->de:4/7 de->b:2/3
d->be:4/9 be->d:100%
e->bd:2/3 bd->e:2/3

要求最小置信度>=50%,所以组合关联规则如下:
a->b a->d a->e e->a c->b b->d d->b b->e e->b c->d d->e e->d
a->de de->a ae->d e->ad ad->e b->de de->b be->d e->bd bd->e