class CollaborativeRec():
def __init__(self, *args, **kwargs):
self.userdata = args[0]
# Information of jobs : key = jobId
self.jobdata = args[1]
self.test = args[2]
#jobMapper must be a (userId,jobId)
self.jobMapper = self.userdata['Jobid']
self.preProcessor = preProcessor()
# function that ineract with Recommend.py
def getRecord(self):
joblist1 = self.user2user()
joblist2 = self.user2company()
joblist = joblist1 + joblist2
for each in joblist2:
print(each)
#filter the list
def user2user(self):
self.similar = []
#need to slice userdata according to the test data
self.userdata, self.test = self.preProcessor.columnSelector(
self.userdata, self.test)
#get user similarity data
self.cos = CosineSimilarity(self.userdata, self.test)
self.similar = self.cos.calculate_cosine()
jobList = []
jobjson = {}
for each in self.similar:
#find the job Id from user Id
jobId = self.jobMapper[each[0]]
#find job in self.job
job = self.jobdata.loc[jobId]
#put job, value in dict
jobdict = job.to_dict()
jobdict['Jobid'] = jobId
jobjsontemp = {'jobdata': jobdict, 'similarity': each[1]}
jobList.append(jobjsontemp)
return jobList
#item to item based
def user2company(self):
self.similar = []
self.jobdata, self.test = self.preProcessor.columnSelector(
self.jobdata, self.test)
#get user to company similarity data
self.cos = CosineSimilarity(self.jobdata, self.test)
self.similar = self.cos.calculate_cosine()
jobList = []
jobjson = {}
for each in self.similar:
job = self.jobdata.loc[each[0]]
jobdict = job.to_dict()
jobdict['Jobid'] = each[0]
jobjsontemp = {'jobdata': jobdict, 'similarity': each[1]}
jobList.append(jobjsontemp)
return jobList
def company2company(self):
self.similar = []
self.jobdata, self.test = self.preProcessor.columnSelector(
self.jobdata, self.test)
#get company to company similarity data
self.cos = CosineSimilarity(self.jobdata, self.test)
#would not work for multiple row
self.similar = self.cos.calculate_cosine()
jobList = []
def user2company(self):
self.similar = []
jobdata = self.jobdata.copy()
testdata = self.test.copy()
self.jobdata_train, self.test_train = self.preProcessor.columnSelector(
jobdata, testdata)
#get user to company similarity data
self.cos = CosineSimilarity(self.jobdata_train, self.test_train)
self.similar = self.cos.calculate_cosine()
return self.similar
def user2company(self):
self.similar = []
self.jobdata, self.test = self.preProcessor.columnSelector(
self.jobdata, self.test)
#get user to company similarity data
self.cos = CosineSimilarity(self.jobdata, self.test)
self.similar = self.cos.calculate_cosine()
jobList = []
jobjson = {}
for each in self.similar:
job = self.jobdata.loc[each[0]]
jobdict = job.to_dict()
jobdict['Jobid'] = each[0]
jobjsontemp = {'jobdata': jobdict, 'similarity': each[1]}
jobList.append(jobjsontemp)
return jobList
class CollaborativeRec():
def __init__(self, *args, **kwargs):
self.userdata = args[0]
# Information of jobs : key = jobId
self.jobdata = args[1]
self.test = args[2]
#jobMapper must be a (userId,jobId)
self.jobMapper = self.userdata['Jobid']
self.preProcessor = preProcessor(self.jobdata, self.test)
# function that ineract with Recommend.py
def getRecord(self):
#joblist1=self.user2user()
joblist2 = self.user2company()
#logic to combine both result
#joblist = joblist1 + joblist2
return self.preProcessor.makeList((joblist2))
def user2user(self):
self.similar = []
#need to slice userdata according to the test data
self.userdata, self.test = self.preProcessor.columnSelector(
self.userdata, self.test)
#get user similarity data
self.cos = CosineSimilarity(self.userdata, self.test)
self.similar = self.cos.calculate_cosine()
#import ipdb; ipdb.set_trace()
jobList = []
for each in self.similar:
#find the job Id from user Id
jobId = self.jobMapper[each[0]]
Similarity = each[1]
jobList.append((jobId, Similarity))
#find job in self.job
return jobList
#item to item based
def user2company(self):
self.similar = []
jobdata = self.jobdata.copy()
testdata = self.test.copy()
self.jobdata_train, self.test_train = self.preProcessor.columnSelector(
jobdata, testdata)
#get user to company similarity data
self.cos = CosineSimilarity(self.jobdata_train, self.test_train)
self.similar = self.cos.calculate_cosine()
return self.similar
def user2user(self):
self.similar = []
#need to slice userdata according to the test data
self.userdata, self.test = self.preProcessor.columnSelector(
self.userdata, self.test)
#get user similarity data
self.cos = CosineSimilarity(self.userdata, self.test)
self.similar = self.cos.calculate_cosine()
#import ipdb; ipdb.set_trace()
jobList = []
for each in self.similar:
#find the job Id from user Id
jobId = self.jobMapper[each[0]]
Similarity = each[1]
jobList.append((jobId, Similarity))
#find job in self.job
return jobList
class CollaborativeRec():
def __init__(self, *args, **kwargs):
self.userdata = args[0]
# Information of jobs : key = jobId
self.jobdata= args[1]
self.test= args[2]
#jobMapper must be a (userId,jobId)
self.jobMapper= self.userdata['Jobid']
self.preProcessor = preProcessor(self.jobdata,self.test)
# function that ineract with Recommend.py
def getRecord(self):
#joblist1=self.user2user()
joblist2=self.user2company()
#logic to combine both result
#joblist = joblist1 + joblist2
return self.preProcessor.makeList((joblist2))
def user2user(self):
self.similar= []
#need to slice userdata according to the test data
self.userdata, self.test= self.preProcessor.columnSelector(self.userdata, self.test)
#get user similarity data
self.cos= CosineSimilarity(self.userdata,self.test)
self.similar = self.cos.calculate_cosine()
#import ipdb; ipdb.set_trace()
jobList= []
for each in self.similar:
#find the job Id from user Id
jobId= self.jobMapper[each[0]]
Similarity= each[1]
jobList.append((jobId,Similarity))
#find job in self.job
return jobList
#item to item based
def user2company(self):
self.similar= []
jobdata= self.jobdata.copy()
testdata= self.test.copy()
self.jobdata_train, self.test_train= self.preProcessor.columnSelector(jobdata,testdata)
#get user to company similarity data
self.cos= CosineSimilarity(self.jobdata_train,self.test_train)
self.similar = self.cos.calculate_cosine()
return self.similar
def user2company(self):
self.similar= []
jobdata= self.jobdata.copy()
testdata= self.test.copy()
self.jobdata_train, self.test_train= self.preProcessor.columnSelector(jobdata,testdata)
#get user to company similarity data
self.cos= CosineSimilarity(self.jobdata_train,self.test_train)
self.similar = self.cos.calculate_cosine()
return self.similar
def company2company(self):
self.similar= []
self.jobdata, self.test= self.preProcessor.columnSelector(self.jobdata, self.test)
#get company to company similarity data
self.cos= CosineSimilarity(self.jobdata,self.test)
#would not work for multiple row
self.similar = self.cos.calculate_cosine()
jobList=[]
# extract maximum from each row
def getRecord(self):
self.similar= []
self.cos= CosineSimilarity(self.userdata,self.test)
self.similar = self.cos.calculate_cosine()
jobList= []
jobjson= {}
for each in self.similar:
#find the job Id from user Id
jobId= self.jobMapper[each[0]]
#find job in self.job
job = self.jobdata.loc[jobId]
#put job, value in dict
jobdict= job.to_dict()
jobdict['Jobid']= jobId
jobjsontemp = {'jobdata':jobdict,'similarity':each[1]}
jobList.append(jobjsontemp)
return jobList
def user2user(self):
self.similar = []
#need to slice userdata according to the test data
self.userdata, self.test = self.preProcessor.columnSelector(
self.userdata, self.test)
#get user similarity data
self.cos = CosineSimilarity(self.userdata, self.test)
self.similar = self.cos.calculate_cosine()
jobList = []
jobjson = {}
for each in self.similar:
#find the job Id from user Id
jobId = self.jobMapper[each[0]]
#find job in self.job
job = self.jobdata.loc[jobId]
#put job, value in dict
jobdict = job.to_dict()
jobdict['Jobid'] = jobId
jobjsontemp = {'jobdata': jobdict, 'similarity': each[1]}
jobList.append(jobjsontemp)
return jobList
def user2user(self):
self.similar= []
#need to slice userdata according to the test data
self.userdata, self.test= self.preProcessor.columnSelector(self.userdata, self.test)
#get user similarity data
self.cos= CosineSimilarity(self.userdata,self.test)
self.similar = self.cos.calculate_cosine()
#import ipdb; ipdb.set_trace()
jobList= []
for each in self.similar:
#find the job Id from user Id
jobId= self.jobMapper[each[0]]
Similarity= each[1]
jobList.append((jobId,Similarity))
#find job in self.job
return jobList
class ContentRec():
def __init__(self, *args, **kwargs):
# Information of jobs : key = jobId
self.jobdata = args[1]
self.test = args[2]
#jobMapper must be a (userId,jobId)
self.preProcessor = preProcessor(self.jobdata)
def company2company(self):
self.similar = []
self.jobdata, self.test = self.preProcessor.columnSelector(
self.jobdata, self.test)
#get company to company similarity data
self.cos = CosineSimilarity(self.jobdata, self.test)
#would not work for multiple row
self.similar = self.cos.calculate_cosine()
jobList = []
class ContentRec():
def __init__(self, *args, **kwargs):
# Information of jobs : key = jobId
self.jobdata= args[1]
self.test= args[2]
#jobMapper must be a (userId,jobId)
self.preProcessor = preProcessor(self.jobdata)
def company2company(self):
self.similar= []
self.jobdata, self.test= self.preProcessor.columnSelector(self.jobdata, self.test)
#get company to company similarity data
self.cos= CosineSimilarity(self.jobdata,self.test)
#would not work for multiple row
self.similar = self.cos.calculate_cosine()
jobList=[]
# extract maximum from each row
class CollaborativeRec():
def __init__(self, *args, **kwargs):
self.userdata = args[0]
# Information of jobs : key = jobId
self.jobdata= args[1]
self.test= args[2]
self.preProcess()
def preProcess(self):
#jobMapper must be a (userId,jobId)
self.jobMapper= self.userdata['Jobid']
self.userdata.drop('Jobid',axis= 1, inplace= True)
order =['experience','php','python','qa','js','level','qualification','age']
self.test = self.test[order]
# If Job is present
#self.test.drop('Jobid',axis= 1, inplace= True)
def getRecord(self):
self.similar= []
self.cos= CosineSimilarity(self.userdata,self.test)
self.similar = self.cos.calculate_cosine()
jobList= []
jobjson= {}
for each in self.similar:
#find the job Id from user Id
jobId= self.jobMapper[each[0]]
#find job in self.job
job = self.jobdata.loc[jobId]
#put job, value in dict
jobdict= job.to_dict()
jobdict['Jobid']= jobId
jobjsontemp = {'jobdata':jobdict,'similarity':each[1]}
jobList.append(jobjsontemp)
return jobList