Learning Apriori Algorithm - in Python

"Machine Learning in Action" is a good book. I've learnt Apriori algorithm successfully. Here is a working Python3 code piece:

#   Load Data
def loadDataSet(path):
return [[1, 3, 4],
[2, 3, 5],
[1, 2, 3, 5],
[2, 5]]

'''
======== Frequent Set Searching ========
'''

#   Create size1 sets
def createC1(dataSet):
C1 = []
#   TODO: list to set maybe good enough
for transaction in dataSet:
for item in transaction:
if not [item] in C1:
C1.append([item])
C1.sort()
return map(frozenset, C1)

# Pruning out all set with support < minSupport
#   D - dataset
#   Ck - candidate sets
#   minSupport - threshold
def scanD(D, Ck, minSupport):
ssCnt = {}
for tid in D:
for can in Ck:
if can.issubset(tid):
if not can in ssCnt: ssCnt[can] = 1
else: ssCnt[can] += 1
numItems = float(len(D))
retList = []
supportData = {}
# Measure support and prune
for key in ssCnt:
support = ssCnt[key] / numItems
if support >= minSupport:
retList.insert(0, key)
supportData[key] = support
return retList, supportData

def aprioriGen(Lk, k): # creates Ck
retList = []
lenLk = len(Lk)
for i in range(lenLk):
for j in range(i + 1, lenLk):
L1 = list(Lk[i])[:k-2]  # [0,1] | [0,2] -> [0,1,2]
L2 = list(Lk[j])[:k-2]
if L1 == L2:
retList.append(Lk[i] | Lk[j])
return retList

def apriori(dataSet, minSupport = 0.5):
# start from size 1
C1 = list(createC1(dataSet))
D = list(map(set, dataSet))
L1, supportData = scanD(D, C1, minSupport)
#
L = [L1]
k = 2
while(len(L[k-2]) > 0):
print ('=Debug= Apriori Size of Last Level', len(L[k-2]))
Ck = aprioriGen(L[k-2], k)
Lk, supK = scanD(D, Ck, minSupport)
supportData.update(supK)
L.append(Lk)
k += 1

return L, supportData

'''
======== Association Rule Searching ========
H: a list of items that could be on the right-hand side of a rule
'''

def calcConf(freqSet, H, supportData, brl, minConf=0.7):
prunedH = []
for conseq in H:
conf = supportData[freqSet] / supportData[freqSet - conseq]
if conf >= minConf:
print (set(freqSet - conseq), '-->', set(conseq), 'conf:', conf * 100, '%')
brl.append((freqSet - conseq, conseq, conf))
prunedH.append(conseq)
return prunedH

def rulesFromConseq(freqSet, H, supportData, brl, minConf = 0.7):
m = len(H[0])
if (len(freqSet) > (m + 1)):
Hmp1 = aprioriGen(H, m + 1) # Gen list of next iteration
Hmp1 = calcConf(freqSet, Hmp1, supportData, brl, minConf) # pruning. pick qualified rules.
if (len(Hmp1) > 1):
rulesFromConseq(freqSet, Hmp1, supportData, brl, minConf) # Continue\Iterate to next level

#   L: a set of freqent itemset; sorted by length
def generateRules(L, supportData, minConf = 0.7):
bigRuleList = []
for i in range(1, len(L)): # from length 2
#print ('Apriori Rule ', i)
for freqSet in L[i]:
H1 = [frozenset([item]) for item in freqSet] # {0,1,2} -> [{0},{1},{2}].
# Build from size 1 on right-hand side
if (i > 1): # length > 2, go level by level
rulesFromConseq(freqSet, H1, supportData, bigRuleList, minConf)
else: # if only 2 items, just prune - the base
calcConf(freqSet, H1, supportData, bigRuleList, minConf)
return bigRuleList