数据挖掘 Apriori Algorithm python实现

该算法主要是处理关联分析的；

大多书上面都会介绍，这里就不赘述了；

dataset=[[1,2,5],[2,4],[2,3],[1,2,4],[1,3],[2,3],[1,3],[1,2,3,5],[1,2,3]]
def init(dataset):
    sset=[]
    for i in dataset:
        for j in i:
            if not [j] in sset:
                sset.append([j])
    sset.sort()
    return list(map(frozenset,sset))

def scan(D,Ck,minsupport):
    # D：数据集；Ck候选集；minS:最小支持度
    cnt={}
    for i in D:
        for j in Ck:
            if j.issubset(i):
                if j not in cnt.keys():cnt[j]=1
                else : cnt[j]+=1
    number=int(len(D))
    Lk=[]#频繁k项集
    supportdata={}
    for item in cnt:
        support=cnt[item]/number
        if support>=minsupport:#大于最小支持度就加入
            Lk.append(item)
        supportdata[item]=support
    return Lk,supportdata

def Link(Lk,k):
    #将频繁k-1项集拼接为候选k项集
    Ck=[]
    length=len(Lk)
    for i in range(length):
        l1=list(Lk[i])[:k-2]
        l1.sort()
        for j in range(i+1,length):
            l2=list(Lk[j])[:k-2]
            l2.sort()
            if l1==l2: Ck.append(Lk[i]|Lk[j])# union
    return Ck

def AprioriAlgo(dataset,minsupport):
    sset=init(dataset)
    L1,supportdata=scan(dataset,sset,minsupport)
    L=[L1]
    k=2
    while(len(L[k-2])>0):
        l1=L[k-2]
        ck=Link(l1,k)
        print("ck: ",ck)
        lk,supk=scan(dataset,ck,minsupport)
        supportdata.update(supk)
        print("lk: ",lk)#频繁k项集
        L.append(lk)
        k+=1
    return L,supportdata

L,supportdata=AprioriAlgo(dataset,minsupport=0.2)