Fun Project: Phone Numbers Validation Tool using NumVerify API

Author: LI LIU (1st yr master’s student at UChicago)

10/26/18: basic framework

10/27/18: add multiple conditions for verifying valid numbers

import pandas as pd

#Read in file
Num = pd.read_excel('PhoneSurvey.xlsx', 
                    names=['ID','Phone','Response','Party','Age'])

Example: Random-genrated Numbers in Philadelphia (Code 267)

Num["Code"] = 267
Num["PN"] = Num["Code"].astype(str) + Num["Phone"].astype(str)
Num.head()

	ID	Phone	Response	Party	Age	Code	PN
0	1	4310454	NaN	NaN	NaN	267	2674310454
1	2	6920338	NaN	NaN	NaN	267	2676920338
2	3	6918070	NaN	NaN	NaN	267	2676918070
3	4	3605305	NaN	NaN	NaN	267	2673605305
4	5	8892229	NaN	NaN	NaN	267	2678892229

Create a free account to get access key

One account is good for to 250 API requests.

Go to https://numverify.com/documentation!

Define the fucntion

import urllib.request, json 
list=[]

#Replace with access key
key = "###"

def validtest(number):
    '''
    Return the indicator list of whether number is valid or not.
    '''
    link = "http://apilayer.net/api/validate?access_key="+key+"&number="+\
            number + "&country_code=US&format=1"
    with urllib.request.urlopen(link) as url:
         data = json.loads(url.read().decode())
            
         #Specify conditions for valid numbers
         if data["valid"] == True and data['carrier'] !='' and data['location'] !='' and data['line_type'] !='':
            list.append(1)
         else:
            list.append(0)
                
    return data

Run the function

for num in Num["PN"]:
    validtest(num)
res = pd.Series(list)
Num["Label"] = res.values

#Counts of valid numbers
sum(list)

60

#Test single number
#validtest("my number")

Num.head()

	ID	Phone	Response	Party	Age	Code	PN	Label
0	1	4310454	NaN	NaN	NaN	267	2674310454	1
1	2	6920338	NaN	NaN	NaN	267	2676920338	0
2	3	6918070	NaN	NaN	NaN	267	2676918070	0
3	4	3605305	NaN	NaN	NaN	267	2673605305	0
4	5	8892229	NaN	NaN	NaN	267	2678892229	1

Save the output

Num.to_excel('data.xlsx')