序列挖掘——SPADE的python实现

def arrange(list,minsup):
sigma=[]
for itemset in list:
for item in itemset:
if item not in sigma:
sigma.append(item)
print(sigma)
dic={}
for name in sigma:
idic={}
for i in range(len(list)):
s=[]
for j in range(len(list[i])):
if name == list[i][j]:
s.append(j)
if s:
idic[i]=s
if len(idic)>=minsup:
dic[name]=idic
return dic

def SequentialPattern(minsup,sequentialPattern,dic):
newdic={}
for sname in sequentialPattern.keys():
for name in dic.keys():
newnumdic={}
for num in sequentialPattern[sname].keys():
if dic[name][num]:
numset=[]
for item in dic[name][num]:
if item>sequentialPattern[sname][num][0]:
numset.append(item)
if numset:
newnumdic[num]=numset
if len(newnumdic)>=minsup:
newdic[sname+name]=newnumdic
return newdic

if name == ‘main’:
list=[[‘C’,‘A’,‘G’,‘A’,‘A’,‘G’,‘T’],[‘T’,‘G’,‘A’,‘C’,‘A’,‘G’],[‘G’,‘A’,‘A’,‘G’,‘T’]]
minsup=3
dic=arrange(list,minsup)
print(dic)
sequentialPattern=SequentialPattern(minsup,dic,dic)
print(sequentialPattern)
sequentialPattern = SequentialPattern(minsup, sequentialPattern, dic)
print(sequentialPattern)
sequentialPattern = SequentialPattern(minsup, sequentialPattern, dic)
print(sequentialPattern)

最后几行明显可以使用递归的这里懒得写了。

本人才疏学浅，如有问题还望及时纠正，谢谢。