DateSafe

 DateSafe: Maintaining Privacy on Tinder

Reason to choose this topic: 

We came across some new headlines that dating apps during COVID-19 experience an increase in the number of users. Our team got fascinated with this news article and discussed the psychological reason for this rise in the new user's graph. For our project, we chose Tinder as the selected dating 

app because of the amount of PII we get from it, which is much more than any other dating app 

such as OkCupid or Bumble. 

Introduction:

In the world of virtual life, you come across many instances where you virtually meet people who are far away from you. You don't even know them. That is the real magic of social media. Everything you 

Whether it be your details or interests, share or post on your social media account is linked. 

After reading this, the first thought will be how can I have my privacy on these social media accounts. 

We will be answering it. We aim to find social media accounts(Facebook or Instagram) from a Tinder profile. This will be possible by processing the collected sample of data from Tinder which is linked from Facebook. Let us show you how a Tinder profile is linked to Facebook.

After login in, Tinder asks various questions regarding your identity. This leads to how secure the user account is. Tinder verifies its user from their life.

The following are the PII that Tinder asks from the user.

Tinder asks the user for their personal information, which will be shown to other users. 

It entirely depends on the user how much he should compromise his privacy or not. 

The psychological fact is that privacy factors are not the same for everybody. 

Tinder requests some basic information of the user from Facebook. That information consist of the 

following information: 

1. Name
2. Email address
3. Location
4. Friend List (this feature is not available in the Indian version of Tinder)
5. Relationship Interests
6. Education history
7. Work History
8. Photos uploaded

We decided to make some of them our PII, which will help us track the user on social media through 

an app or Facebook query manager.

Detecting PII from Tinder Profile:

We made a python code that will be functioning as a form, which will ask the user about the profile's PII.

The PII that we collected are:

1. Name
2. City
3. Institution Name
4. Workplace      

Building Facebook Search Link Generator:

Analysing the URL changes according to every state, country, city. 

As we know that Facebook has stopped the search queries to work. So, our team has observed some of the search URL’s patterns which will be helping us to make a code that can give Facebook URL as an outcome, and the input will be a FOR loop which will be reading the PII from our stored CSV file. 

Like, search a person with the name = Manish Kumari

https://www.facebook.com/search/people/?q=manish%20kumari

The highlighted text is fixed in the case of user (name) search.

https://www.facebook.com/search/people/?q=manish%20kumari

Similarly, for other filters to we analysed some common URL identifying text.

Another example will be what is we know that Manish Kumari lives in Delhi

https://www.facebook.com/search/people?q=manish%20kumari&filters=eyJjaXR5OjAiOiJ7XCJuYW1lXCI6XCJ1c2Vyc19sb2NhdGlvblwiLFwiYXJnc1wiOlwiMTAyMTYxOTEzMTU4MjA3XCJ9In0%3D

So, here & implies that we have added another filter, and then the furthermore part of the URL is for Delhi (code URL).

We will be making a file in excel file to save all the Cities and their Facebook code. 

Like, for Japan it is 

eyJjaXR5OjAiOiJ7XCJuYW1lXCI6XCJ1c2Vyc19sb2NhdGlvblwiLFwiYXJnc1wiOlwiMTA2NTE0MDA2MDUzMjUwXCJ9In0

(Keeping TinderGold in mind)

When compared with Delhi’s URL 

eyJjaXR5OjAiOiJ7XCJuYW1lXCI6XCJ1c2Vyc19sb2NhdGlvblwiLFwiYXJnc1wiOlwiMTAyMTYxOTEzMTU4MjA3XCJ9In0

The highlighted part of the URL is identical for every city or country.

The challenge is that the URL of the Facebook filter changes while allotting it two filters. 

Like, if we have a filter of Delhi having a URL (eyJjaXR5OjAiOiJ7XCJuYW1lXCI6XCJ1c2Vyc19sb2NhdGlvblwiLFwiYXJnc1wiOlwiMTAyMTYxOTEzMTU4MjA3XCJ9In0)

And a filter of IIIT Delhi(eyJzY2hvb2w6MCI6IntcIm5hbWVcIjpcInVzZXJzX3NjaG9vbFwiLFwiYXJnc1wiOlwiMjMxMzg3OTkwMjExNTcyXCJ9In0%3D)

So their combination is different. So, it is time-consuming for us to mention all the filters and combine them, like a person who lives in Delhi and has IIT Delhi as its institute, then we have to make its combination and store it in a CSV file. Similarly for all cities and institutes and workplaces.

Making Data Set:

We made a Data set, which consists of all the cities in India. All the educational institutes, whether it be school or college. All the workplaces in India. So, this dataset is made manually, which took hours, but it's still under process. For some cities and Institutes which were famous, we made filter code python for that. 

Work CSV

                                                                        Education CSV

    Our Dataset is massive, and we have covered all the Institutes, Cities, States, Workplaces within             India.

This is our python code. *NOT COMPLETE CODE*

Working:

Taking a profile.

Asking for PII

Copying the URL generated by our python code and opening that URL.

Observation: 

We were able to find the following ratio of the users.

Girls: 47 out of 62 profiles 

Boys: 22 out of 44 profiles 

The ratio of boys is less as compared to the girls found. The primary reason is that girls tend to make themselves updated and more active on social media than boys.

                        Profiles Collected                                                  Profiles Found

Privacy Maintainance on Tinder:

The user can maintain their privacy by following these two rules.

1. Pre-registration 
2. After-Registration

In Pre-registration, the user has to provide their PII to Tinder for profile making purpose. So, the user can even give their nicknames their wrong age. 

Providing with their school or institution is another mistake that user’s commit.

In After-Registration, the user needs to follow these norms.

There can be a case where someone does not want anybody to track them on social media; they should get their eyes on this.

1. Do not link the Instagram if you do not want to get recognized on social platforms easily.
2. Do not put social media handles in bio to avoid unwanted texts and requests.
3. Do think about your privacy when you fill in the details that Tinder asks you to fill before setting up your profile. (Pre-Registration)
4. Try not to enter your surname. That gives the user a solid chance to get your social media handle.
5. Do not use the same profile pic(on Facebook and Instagram) that you have uploaded on Tinder.

Psychological Inferences:

We conducted a survey to analyse user's mindset and problems. 

  

1. Our target audience is the people who found some profile attractive and that user wants him or her to approach directly. Rather than waiting for him or her to the right, wipe them.
2. Users will use our code to check whether the profile they are interested in is genuine or not. If the profile has a match on Facebook and that too with the same PII (including education history), then there is a very significantly less chance of that profile to be fake. 
3. Most people prefer not showing themselves, or their social media handles on their Tinder profile. This can be that the user does not want to be recognized by someone known, or the user does not wish to the opposite person to see their mutual friends on social media platforms. It is sometimes embarrassing for some people to commit publicly that they are on a dating site.

Team Members:

Parth Singh 2018356

Yatharth Chauhan 2018373

Anil Kumar 2018328

Shekhar 2018365

Manish Kumar Kushwaha 2018156

Nikhil Kumar 2018163

Nikhil Dahiya 2018057