Ben-Gurion Team Develops Facebook Phony Friend Detector
Facebook is a “friendly” place; the purpose of Facebook is to share information with friends. For many users, especially young ones, amassing friends is a major activity, and it isn’t unusual to find teens with 500, 1,000, or even more friends.
But as recent scandals have shown, social networks can be magnets for not only friends, but fiends, such as pedophiles, child molesters, and other abominable types. Until now, it’s been very difficult, if not impossible, to ferret the phonies out. But a new application from Ben-Gurion University researchers may help more users – and the parents of users – keep undesirable false friends off their connection list.
- Pinvolve App Turns Facebook Pages Into Pinboards
- Happy Bday: Facebook App To Help You Find The Right Gift
The team — Michael Fire, Dima Kagan, Aviad Elishar, and Yuval Elovici of the Telekom Innovation Laboratories and Information Systems Engineering Department of Ben-Gurion University – developed a Facebook app called Social Privacy Protector (SPP), based on research they did on social networks in general, and Facebook in particular. The project, Fire told The Times of Israel, was originally developed as a method to detect pedophiles, but could also be used to detect “friends” with agendas that are different than those they set out on their Facebook pages.
“The amount of personal information each user exposes on social networks such as Facebook is staggering,” said Fire. “Recent research in the area of social networking evaluated that many Facebook users exposed personal information. Due to the many security concerns regarding online personal exposure, we developed SPP, a software which aims to improve the security and privacy of Facebook users.”
The software is available as a Facebook app for all browsers, and as an add-on for Firefox, giving users much firmer control over their Facebook privacy settings.
The software contains three protection layers which improve user privacy, said Fire, “identifying a user’s friends who might pose a threat, and then restricting the ‘friend’s’ exposure to the user’s personal information. The second layer is an expansion of Facebook’s basic privacy settings based on different types of social network usage profiles. The third layer alerts the user about the number of installed applications on their Facebook profile which have access to their private information.”
SPP does deep analysis of a user’s connections and friends, determining what organic groups they belong to, how often they communicate, the nature of their posts, who responds to them, and other factors. Users get a score based on the strength of the connection between the user and their friends. The application estimates the strength of each connection by calculating the number of common friends between the user and their friend, the number of pictures and videos the user and their friend were tagged in together, the number of groups the user and their friend were both members in, and the number of messages passed between the user and their friend.
When it’s done analyzing, SPP gives a “connectedness score” to each of a user’s contacts. The lower the score, the more “suspicious” the connection. The app lists the connections that scored low, and recommends defriending them.
Does it work? Most emphatically, said Fire. Before moving into Facebook, social networking’s “big leagues,” the team tried out its algorithms on several smaller networks. “During the course of our research, we encountered a profile which appeared in several social networks, was very active, and had many friends. Only through the use of ‘photo watermarks’ were we able to uncover the fact that the picture actually belongs to a different user in a different country.” The profile was a fake that was good enough to “pass” — and heaven knows who, or what, was really behind that phony photo.
“Facebook is evaluating the app, and we expect to get the add-on in the Firefox app site in the coming weeks.” (Currently, it can be downloaded here).
Is there a chance Facebook could turn the app down, saying that it is not appropriate for users? Maybe, but it’s unlikely, said Fire. “Our app really does the opposite of what Facebook wants. The service encourages connecting with as many people as possible, while we advocate limiting users, and have, for the first time, provided an algorithm to scientifically determine who to remove from friend lists. But we believe that the app will be approved,” said Fire.