def computeAlertsChunk(chunkID):
#########################################################################################################
############-----------------Creating a connection to output mongodb
#########################################################################################################
tablename = 'MonthlyMsgQueue'
monconn_recommendations = MongoConnect(tablename,
host='localhost',
database='mailer_monthly')
print 'Chunk:', chunkID, 'initiated at:', time.ctime()
ifile = open('CompanyNames.csv', 'r')
reader = csv.reader(ifile)
company_dict = {}
for row in reader:
company_dict[row[0]] = row[1]
#########################################################################################################
############-----------------Fetch the user data from the database
#########################################################################################################
tablename = "candidates_processed"
monconn_users = MongoConnect(tablename,
host='localhost',
database='mailer_monthly')
mongo_users_cur = monconn_users.getCursor()
myCondition = {'p': chunkID}
#myCondition = {}
users = monconn_users.loadFromTable(myCondition)
#print "Fetching the users data from Mongodb....completed for ChunkID:",chunkID
#########################################################################################################
############-----------------Loop to generate recommendations and save in Mongo
#########################################################################################################
count = 0
for user in users:
#########################################################################################################
############-----------------Extracting the user details
#########################################################################################################
count += 1
user_ctc = user['user_ctc']
user_exp = user['user_experience']
user_id = user['user_id']
user_email = user['user_email']
user_bow = user['user_bow']['bow']
user_current_time = datetime.datetime.now()
user_jobtitle = user['user_jobtitle']
user_lastlogin = user['user_lastlogin']
user_phone = user['user_phone']
user_gender = user['user_gender']
user_current_company = user['user_current_company']
user_functionalarea_id = user['user_functionalarea_id']
user_lastmodified = user['user_lastmodified']
user_fullname = user['user_fullname']
user_phone_verified = user['user_phone_verified']
user_location_id = user['user_location_id']
user_ctc_id = user['user_ctc_id']
user_highest_qual = user['user_highest_qual']
user_edu_special = user['user_edu_special']
user_email_verified = user['user_email_verified']
user_spam_status = user['user_spam_status']
user_bounce_status = user['user_bounce_status']
user_email_alert_status = user['user_email_alert_status']
user_functionalarea = user['user_functionalarea']
user_industry = user['user_industry']
user_jobtitle = user['user_jobtitle']
user_profiletitle = user['user_profiletitle']
user_edom = user['user_edom']
user_industry = user['user_industry']
user_skills = user['user_skills']
user_profiletitle = user['user_profiletitle']
user_pid = user['p']
user_firstname = user_fullname.split(" ")[0]
lsi_user = lsiModel[tfIdfModel[user_bow]]
simScrChunk = index[lsi_user]
sortingExcelSheetList = []
for (jobIntIndex, lsiCosine) in simScrChunk:
if lsiCosine < 0.18:
continue
#########################################################################################################
############-----------------Loading the Jobs Data
#########################################################################################################
job = jobIntIdToJobDict[jobIntIndex]
jobid = job['job_id']
job_title = job['job_title']
job_skills = job['job_skills']
job_minsal = job['job_minsal']
job_maxsal = job['job_maxsal']
job_minexp = job['job_minexp']
job_maxexp = job['job_maxexp']
job_bow = job['job_bow']['bow']
job_accounttype = job['job_accounttype']
job_flag = job['job_flag']
job_companyname = job['job_company_name']
job_companyid = job['job_company_id']
#########################################################################################################
############-----------------Calculating the CTC and Experience Match Scores
#########################################################################################################
ctc_match_score = CTCMatchScore(job_minsal, job_maxsal,
user_ctc).CTCMatchScore()
exp_match_score = ExpMatchScore(job_minexp, job_maxexp,
user_exp).ExpMatchScore()
paid_boost = PaidBoostScore(job_flag,
job_accounttype).PaidBoostScore()
#########################################################################################################
############-----------------Calculating the City Score between a candidate and a job
#########################################################################################################
if ctc_match_score == 1 and exp_match_score == 1:
jobid = job['job_id']
try:
job_city = job['job_location']
except:
job_city = 'Delhi'
try:
user_city = user['user_location']
except:
user_city = 'Delhi'
#print user_city, job_city
try:
user_city_list = user_city.lower().replace(
'other', '').strip().split(',')
user_city_list = [x.strip() for x in user_city_list]
except:
user_city_list = ['']
try:
job_city_list = job_city.lower().replace(
'other', '').strip().split(',')
job_city_list = [x.strip() for x in job_city_list]
except:
job_city_list = ['']
#print user_city_list, job_city_list
try:
cityScore = cm.getCityScore(user_city_list, job_city_list)
except:
cityScore = 0
#########################################################################################################
############-----------------Calculating the overall match score and appending the details to the list
############-----------------based on job's published date
#########################################################################################################
overallMatchScore = getOverallMatchScore(
lsiCosine, cityScore, paid_boost)
s = (user_id, user_email, jobid, overallMatchScore, job_title,
job_skills, job_minsal, job_maxsal, job_minexp,
job_maxexp, job_companyid)
sortingExcelSheetList.append(s)
else:
continue
##############################################################################################################
############-----------------Finding the top 10 Jobs based on Overall Score
##############################################################################################################
topN = 30
sortingExcelSheetListTopNJobs = heapq.nlargest(topN,
sortingExcelSheetList,
key=lambda x: x[3])
#pprint(sortingExcelSheetListTopNJobs)
jobs2bsent = []
company_ids = []
cosine_score = []
for (user_id, user_email, jobid, overallMatchScore, job_title,
job_skills, job_minsal, job_maxsal, job_minexp, job_maxexp,
job_companyid) in sortingExcelSheetListTopNJobs:
#print (userid, jobid, lsiCosine, job_title, job_skills, job_minsal, job_maxsal, job_minexp, job_maxexp)
if job_companyid not in company_ids:
company_ids.append(job_companyid)
jobs2bsent.append(int(jobid))
cosine_score.append(round(overallMatchScore, 2))
else:
if company_ids.count(job_companyid) < 3:
company_ids.append(job_companyid)
jobs2bsent.append(int(jobid))
cosine_score.append(round(overallMatchScore, 2))
else:
pass
if len(jobs2bsent) >= 10:
break
else:
pass
#print user_id
#print company_ids
#print jobs2bsent
companies = []
#print company_ids
for comp_id in company_dict.keys():
if int(comp_id) in company_ids:
companies.append(company_dict[comp_id])
#print companies
#print "Hello"
else:
pass
##############################################################################################################
############-----------------Creating Subject Line for a candidate
############################################################################################################## \
if len(companies) != 0:
try:
user_subject = user_firstname + ": " + ', '.join(
companies
) + " and other top company jobs matching your profile"
#print user_subject
except Exception as e:
pass
else:
try:
if user_functionalarea == "Fresher (No Experience)":
user_subject = user_firstname + ", don't miss out on these new jobs"
else:
user_subject = user_firstname + ", new " + user_functionalarea.replace(
' /', ',') + " jobs for you"
#print user_subject
except Exception as e:
user_subject = user_firstname + ", don't miss out on these new jobs"
##############################################################################################################
############-----------------Creating a document to be saved in mongo collection
##############################################################################################################
document = {
"c": user_id,
"_id": user_email,
"m": user_phone,
"te": user_exp,
"cr": user_jobtitle,
"g": user_gender,
"cc": user_current_company,
"fa": user_functionalarea,
"faid": user_functionalarea_id,
"pd": user_lastmodified,
"fn": user_fullname,
"cpv": user_phone_verified,
"sCLID": user_location_id,
"sASID": user_ctc_id,
"eq": user_highest_qual,
"es": user_edu_special,
"ev": user_email_verified,
"ll": user_lastlogin,
"sal": user_ctc,
"cosine": cosine_score,
"edom": user_edom,
"t": user_current_time,
"mj": jobs2bsent,
"bj": [],
"oj": [],
"pid": user_pid,
"s": False,
"sub": user_subject
}
##############################################################################################################
############-----------------Dumping the document in mongo collection if recommendations were generated
##############################################################################################################
if len(jobs2bsent) > 0:
monconn_recommendations.saveToTable(document)
#print 'Chunk:', chunkID, 'processed in:', time.ctime()
monconn_recommendations.close()
def computeAlertsChunk(chunkID):
#########################################################################################################
############-----------------Creating a connection to output mongodb
#########################################################################################################
tablename = 'JobSuggestions'
monconn_recommendations = MongoConnect(tablename,
host='localhost',
database='similar_jobs_onsite')
print 'Chunk:', chunkID, 'initiated at:', time.ctime()
#########################################################################################################
############-----------------Fetch the 3 month jobs data from mongo
#########################################################################################################
tablename = "active_jobs_dump"
monconn_jobs_1 = MongoConnect(tablename,
host='localhost',
database='similar_jobs_onsite')
mongo_jobs_1_cur = monconn_jobs_1.getCursor()
myCondition = {'pid': chunkID}
jobs_1 = monconn_jobs_1.loadFromTable(myCondition)
#########################################################################################################
############-----------------Calculating the overall score of a 3month jobs based on cosine,ctc,
############-----------------experience,city scores for each 1month Job
#########################################################################################################
count = 0
for job_1 in jobs_1:
count += 1
jobid_1 = job_1['job_id']
job_title_1 = job_1['job_title']
job_skills_1 = job_1['job_skills']
job_minsal_1 = job_1['job_minsal']
job_maxsal_1 = job_1['job_maxsal']
job_minexp_1 = job_1['job_minexp']
job_maxexp_1 = job_1['job_maxexp']
job_bow_1 = job_1['job_bow']['bow']
job_index_1 = job_1['job_index']
lsi_job_1 = lsiModel[tfIdfModel[job_bow_1]]
simScrChunk = index[lsi_job_1]
sortingExcelSheetList = []
for (jobIntIndex, lsiCosine) in simScrChunk:
job = jobIntIdToJobDict[jobIntIndex]
jobid = job['job_id']
job_title = job['job_title']
job_skills = job['job_skills']
job_minsal = job['job_minsal']
job_maxsal = job['job_maxsal']
job_minexp = job['job_minexp']
job_maxexp = job['job_maxexp']
job_bow = job['job_bow']['bow']
job_index = job['job_index']
job_company_id = job['job_company_id']
#########################################################################################################
############-----------------Calculating the CTC and Experience and City Match Scores
#########################################################################################################
ctc_match = CTCMatchScore(job_minsal_1, job_maxsal_1, job_minsal,
job_maxsal)
ctc_match_score = ctc_match.CTCMatchScore()
exp_match_score = ExpMatchScore(job_minexp_1, job_maxexp_1,
job_minexp,
job_maxexp).ExpMatchScore()
paid_boost = 0
if ctc_match_score == 1 and exp_match_score == 1:
if jobid != jobid_1:
try:
job_city_1 = job_1['job_location']
except:
job_city_1 = ["Delhi"]
try:
job_city = job['job_location']
except:
job_city = ["Delhi"]
#lsiCosine = getLSICosine(user_bow, job_bow).getLSICosine()
try:
cityScore = cm.getCityScore(job_city_1, job_city)
except:
cityScore = 0
overallMatchScore = getOverallMatchScore(
lsiCosine, cityScore, paid_boost)
s = (jobid_1, job_index_1, jobid, job_index,
overallMatchScore, job_company_id)
sortingExcelSheetList.append(s)
else:
continue
else:
continue
#########################################################################################################
############-----------------Finding the top 10 Jobs based on overall sccore
#########################################################################################################
topN = 30
sortingExcelSheetListTopNJobs = heapq.nlargest(topN,
sortingExcelSheetList,
key=lambda x: x[4])
jobs2bsent = []
company_ids = []
for (jobid_1, job_index_1, jobid, job_index, overallMatchScore,
job_company_id) in sortingExcelSheetListTopNJobs:
if job_company_id not in company_ids:
company_ids.append(job_company_id)
jobs2bsent.append(int(jobid))
else:
if company_ids.count(job_company_id) < 2:
company_ids.append(job_company_id)
jobs2bsent.append(int(jobid))
else:
pass
if len(jobs2bsent) >= 10:
break
else:
pass
##############################################################################################################
############-----------------Creating a document to be saved in mongo collection
############################################################################################################## \
document = {
'_id': jobid_1,
'sj': jobs2bsent,
'sjlen': len(jobs2bsent),
'lud': datetime.datetime.now()
}
##############################################################################################################
############-----------------Dumping the document in mongo collection if recommendations were generated
##############################################################################################################
monconn_recommendations.saveToTable(document)
monconn_recommendations.close()
def computeAlertsChunk(chunkID):
#########################################################################################################
############-----------------Creating a connection to output mongodb
#########################################################################################################
tablename = 'DailyMsgQueue'
monconn_recommendations = MongoConnect(tablename,
host='localhost',
database='mailer_daily_midout')
print 'Chunk:', chunkID, 'initiated at:', time.ctime()
#########################################################################################################
############-----------------Fetch the user data from the database
#########################################################################################################
tablename = "candidates_processed_midout"
monconn_users = MongoConnect(tablename,
host='localhost',
database='Midout_Mailers')
mongo_users_cur = monconn_users.getCursor()
myCondition = {'pid': chunkID}
users = monconn_users.loadFromTable(myCondition)
#########################################################################################################
############-----------------Loop to generate recommendations and save in Mongo
#########################################################################################################
count = 0
for user in users:
count += 1
#########################################################################################################
############-----------------Extracting the user details
#########################################################################################################
_id = user['user_email']
user_ctc = user['user_ctc']
user_exp = user['user_experience']
user_id = user['user_id']
user_email = user['user_email']
user_bow = user['user_bow']['bow']
user_current_time = datetime.datetime.now()
user_countrycode = user.get('user_countrycode', '')
user_pid = user['pid']
user_phone_number = "+" + str(user_countrycode) + "-" + str(
user.get('user_phone_number', ''))
user_firstname = user.get('user_firstname', '')
user_lastname = user.get('user_lastname', '')
user_fullname = user_firstname + " " + user_lastname
if len(user_fullname) < 3:
user_fullname = "Candidate"
try:
user_registration_start_date = rfc3339(user.get(
'user_registration_start_date', ''),
utc=True)
except:
user_registration_start_date = ""
lsi_user = lsiModel[tfIdfModel[user_bow]]
simScrChunk = index[lsi_user]
sortingExcelSheetList = []
for (jobIntIndex, lsiCosine) in simScrChunk:
if lsiCosine < 0.18:
continue
#########################################################################################################
############-----------------Loading the Jobs Data
#########################################################################################################
job = jobIntIdToJobDict[jobIntIndex]
jobid = job['job_id']
job_title = job['job_title']
job_skills = job['job_skills']
job_minsal = job['job_minsal']
job_maxsal = job['job_maxsal']
job_minexp = job['job_minexp']
job_maxexp = job['job_maxexp']
job_bow = job['job_bow']['bow']
#########################################################################################################
############-----------------Calculating the CTC and Experience Match Scores
#########################################################################################################
ctc_match_score = CTCMatchScore(job_minsal, job_maxsal,
user_ctc).CTCMatchScore()
exp_match_score = ExpMatchScore(job_minexp, job_maxexp,
user_exp).ExpMatchScore()
#########################################################################################################
############-----------------Calculating the City Score between a candidate and a job
#########################################################################################################
if ctc_match_score == 1 and exp_match_score == 1:
jobid = job['job_id']
#lsiCosine = getLSICosine(user_bow, job_bow).getLSICosine()
#City Score
try:
job_city = job['job_location']
except:
job_city = 'Delhi'
try:
user_city = user['user_location']
except:
user_city = 'Delhi'
#print user_city, job_city
try:
user_city_list = user_city.lower().replace(
'other', '').strip().split(',')
user_city_list = [x.strip() for x in user_city_list]
except:
user_city_list = ['']
try:
job_city_list = job_city.lower().replace(
'other', '').strip().split(',')
job_city_list = [x.strip() for x in job_city_list]
except:
job_city_list = ['']
#print user_city_list, job_city_list
try:
cityScore = cm.getCityScore(user_city_list, job_city_list)
except:
cityScore = 0
#if cityScore == 0:
#count = count +1
#print user_city_list, job_city_list, cityScore
paidboost = 0
#########################################################################################################
############-----------------Calculating the overall match score
#########################################################################################################
overallMatchScore = getOverallMatchScore(
lsiCosine, cityScore, paidboost)
s = (user_id, user_email, jobid, overallMatchScore, job_title,
job_skills, job_minsal, job_maxsal, job_minexp,
job_maxexp)
sortingExcelSheetList.append(s)
else:
continue
##############################################################################################################
############-----------------Finding the top 10 Jobs based on Overall Score
##############################################################################################################
topN = 10
sortingExcelSheetListTopNJobs = heapq.nlargest(topN,
sortingExcelSheetList,
key=lambda x: x[3])
jobs2bsent = []
for (user_id, user_email, jobid, overallMatchScore, job_title,
job_skills, job_minsal, job_maxsal, job_minexp,
job_maxexp) in sortingExcelSheetListTopNJobs:
jobs2bsent.append(int(jobid))
##############################################################################################################
############-----------------Creating a document to be saved in mongo collection
##############################################################################################################
document = {"_id":user_email, \
"fn" : user_fullname,\
"c": user_id,\
"mj":jobs2bsent, \
"oj":[], \
"s": False ,\
"t": user_current_time ,\
"pid": user_pid,\
"m": user_phone_number, \
"pd":user_registration_start_date
}
##############################################################################################################
############-----------------Dumping the document in mongo collection if recommendations were generated
##############################################################################################################
if len(jobs2bsent) > 0:
monconn_recommendations.saveToTable(document)
monconn_recommendations.close()
def computeAlertsChunk(chunkID):
#########################################
'''Creating a connection to output mongodb'''
#########################################
tablename = 'WeeklyMsgQueue'
monconn_recommendations = MongoConnect(tablename,
host='localhost',
database='mailer_weekly')
print 'Chunk:', chunkID, 'initiated at:', time.ctime()
#################################################
'''Fetch the user data from the database'''
#################################################
#print "Fetching the users data from Mongodb for ChunkID:",chunkID
#tablename="candidates_processed"
tablename = "candidates_processed"
monconn_users = MongoConnect(tablename,
host='localhost',
database='mailer_weekly')
mongo_users_cur = monconn_users.getCursor()
myCondition = {'p': chunkID}
#myCondition = {}
users = monconn_users.loadFromTable(myCondition)
#print "Fetching the users data from Mongodb....completed for ChunkID:",chunkID
##################################################################
'''Get the top 10 matching jobs based on cosine for each candidate'''
##################################################################
count = 0
for user in users:
count += 1
user_ctc = user['user_ctc']
user_exp = user['user_experience']
user_id = user['user_id']
user_email = user['user_email']
user_bow = user['user_bow']['bow']
user_current_time = datetime.datetime.now()
user_jobtitle = user['user_jobtitle']
user_lastlogin = user['user_lastlogin']
user_phone = user['user_phone']
user_gender = user['user_gender']
user_current_company = user['user_current_company']
user_functionalarea_id = user['user_functionalarea_id']
user_lastmodified = user['user_lastmodified']
user_fullname = user['user_fullname']
user_phone_verified = user['user_phone_verified']
user_location_id = user['user_location_id']
user_ctc_id = user['user_ctc_id']
user_highest_qual = user['user_highest_qual']
user_edu_special = user['user_edu_special']
user_email_verified = user['user_email_verified']
user_spam_status = user['user_spam_status']
user_bounce_status = user['user_bounce_status']
user_email_alert_status = user['user_email_alert_status']
user_functionalarea = user['user_functionalarea']
user_industry = user['user_industry']
user_jobtitle = user['user_jobtitle']
user_profiletitle = user['user_profiletitle']
user_edom = user['user_edom']
user_industry = user['user_industry']
user_skills = user['user_skills']
user_profiletitle = user['user_profiletitle']
user_pid = user['p']
lsi_user = lsiModel[tfIdfModel[user_bow]]
simScrChunk = index[lsi_user]
sortingExcelSheetList = []
for (jobIntIndex, lsiCosine) in simScrChunk:
if lsiCosine < 0.18:
continue
job = jobIntIdToJobDict[jobIntIndex]
jobid = job['job_id']
job_title = job['job_title']
job_skills = job['job_skills']
job_minsal = job['job_minsal']
job_maxsal = job['job_maxsal']
job_minexp = job['job_minexp']
job_maxexp = job['job_maxexp']
job_bow = job['job_bow']['bow']
job_accounttype = job['job_accounttype']
job_flag = job['job_flag']
#######################################################
''' Calculating the CTC and Experience Match Scores'''
#######################################################
ctc_match_score = CTCMatchScore(job_minsal, job_maxsal,
user_ctc).CTCMatchScore()
exp_match_score = ExpMatchScore(job_minexp, job_maxexp,
user_exp).ExpMatchScore()
paid_boost = PaidBoostScore(job_flag,
job_accounttype).PaidBoostScore()
#ctc_match_score = 1
#exp_match_score = 1
paid_boost = 0
#######################################################
''' For Low earning desperate guy uncomment this '''
#######################################################
'''
if (1 + user_ctc)/(1 + user_exp) < 0.3:
ctc_match_score = 1
exp_match_score = 1
'''
if ctc_match_score == 1 and exp_match_score == 1:
jobid = job['job_id']
try:
job_city = job['job_location']
except:
job_city = 'Delhi'
try:
user_city = user['user_location']
except:
user_city = 'Delhi'
#print user_city, job_city
try:
user_city_list = user_city.lower().replace(
'other', '').strip().split(',')
user_city_list = [x.strip() for x in user_city_list]
except:
user_city_list = ['']
try:
job_city_list = job_city.lower().replace(
'other', '').strip().split(',')
job_city_list = [x.strip() for x in job_city_list]
except:
job_city_list = ['']
#print user_city_list, job_city_list
try:
cityScore = cm.getCityScore(user_city_list, job_city_list)
except:
cityScore = 0
#if cityScore == 0:
#count = count +1
#print user_city_list, job_city_list, cityScore
overallMatchScore = getOverallMatchScore(
lsiCosine, cityScore, paid_boost)
s = (user_id, user_email, jobid, overallMatchScore, job_title,
job_skills, job_minsal, job_maxsal, job_minexp,
job_maxexp)
sortingExcelSheetList.append(s)
else:
continue
#################################
'''Finding the top 10 Jobs'''
#################################
topN = 10
sortingExcelSheetListTopNJobs = heapq.nlargest(topN,
sortingExcelSheetList,
key=lambda x: x[3])
#pprint(sortingExcelSheetListTopNJobs)
jobs2bsent = []
for (user_id, user_email, jobid, overallMatchScore, job_title,
job_skills, job_minsal, job_maxsal, job_minexp,
job_maxexp) in sortingExcelSheetListTopNJobs:
#print (userid, jobid, lsiCosine, job_title, job_skills, job_minsal, job_maxsal, job_minexp, job_maxexp)
jobs2bsent.append(int(jobid))
document = {
"c": user_id,
"_id": user_email,
"m": user_phone,
"te": user_exp,
"cr": user_jobtitle,
"g": user_gender,
"cc": user_current_company,
"fa": user_functionalarea,
"faid": user_functionalarea_id,
"pd": user_lastmodified,
"fn": user_fullname,
"cpv": user_phone_verified,
"sCLID": user_location_id,
"sASID": user_ctc_id,
"eq": user_highest_qual,
"es": user_edu_special,
"ev": user_email_verified,
"ll": user_lastlogin,
"sal": user_ctc,
"edom": user_edom,
"t": user_current_time,
"mj": jobs2bsent,
"bj": [],
"oj": [],
"pid": user_pid,
"s": False
}
if len(jobs2bsent) > 0:
monconn_recommendations.saveToTable(document)
#print 'Chunk:', chunkID, 'processed in:', time.ctime()
monconn_recommendations.close()