Private Data Mining
I’ve got some good news for everybody who dreamed of mining data like the big agencies do. The key to this data is WhatsApp. Whatsapp is the leader of mobil instant messaging in Germany. According to the “Netzoekonom” there are 32 million germans who use Whatsapp. This is a dizzying number when you compare it to the 80 million citizens in Germany. Facebook has 27,38 million users in Germany. This is a important fact, because you can query with the help of WhatsApp and Facebook millions of names with phonenumbers and profile-pictures.
Let me explain how to do this:
You need for this some phonenumbers, some IP-Adresses and Yowsup. Yowsup provides an unofficial API for WhatsApp. So you can with Yowsup the same things as with WhatsApp. The difference is: You can automatize this functions.
In Germany we have different prefix numbers for mobile phone numbers(+49 is the country code for Germany change this to 0 if you are living in Germany):
+49151,+49161,+49171,…
The mobile phonenumber-ranges are from +4915X up to +4917X. After this prefix numbers we have a suffix-number with 7 or 8 digits. With this knowledge we can simple bruteforce the phone numbers and send queries via Yowsup to the WhatsApp server. So we can get from the Server the following information:
- Is this phonenumber registered at WhatsApp?
- Is the profil-picture public? In case of ‘yes’ we can download it.
- Is the Online-Activity public? In case of ‘yes’ we can query if the user is online or not.
- Is the status public? In case of ‘yes’ we can query the status.
- Is the profile-name public? In case of ‘yes’ we can query the name.
Let’s summarize this: We can gather thousands of profile-pictures with phonenumbers. But thats not enough. How can we use this information to gather more information?
We can use Facebook! The Facebook-Messenger asks the user for connecting the phone number with the Facebook-Account. We can use this feature to link phonenumbers to full names, pictures, hobbies, friends, relationships, sexual preference, political preference etc. Facebook provides a login via email and phonenumber. We can use this loginform:
https://www.facebook.com/login/identify?ctx=login
Just insert the phonenumber with countrycode without “+” at the beginning and you get the full name with profile picture if the profile is public. With this name we can use google for more information about the person.
But how can we automatize this query? We can use curl to query the login-form. But I haven’t tried it. Maybe we must solve solve captcha for more quries. If so we can generate a cookie for this with solving one captcha. We can use this cookie with curl too.