How Facebook Can Learn To Properly Collect Big Data and Be Less Invasive

Over 800 million people worldwide use the social media website Facebook. Unsurprisingly, the social media site been the subject of much debate in recent years because personal information has been exposed to many advertising and Internet tracking companies.

Now, the research group Cryptographic Systems at the Multimodal Computing and Interaction (MMCI) in Saarbrucken, Germany, has developed a novel cryptographic method that has the ability to collect data while protecting the privacy of the user. The research group leaders presented their method at Saarland University.

A user's privacy is threatened in two ways during data aggregation, which include where and how the data is aggregated and the publishing of the data. In order to solve both of these issues, the computer scientists of the Cryptographic Systems research group split up the requested information and sent parts of it to the servers that perform the computation. By splitting up the data, the servers are unable to identify personal information because it was removed from the cluster of original information.

"Many website providers are able to collect data, but only a few manage to do so without invading users' privacy," said Aniket Kate, leader of the Cryptographic Systems research group, in a news release.

When website owners are interested in gathering information about their visitors, such as age and gender, they may have access to sensitive personal information such as what other websites that person visits. This may allow the website owners to reconstruct detailed profiles of each individual. But this new method will only provide selective information where privacy will be honored.

Cryptographic Systems has already implemented their method and it has proved to be successful. The computation is still fast, servers still receive their information in a timely manner, and the information is still as accurate as possible.

To read more about the new data-protected method, check out Science Daily research.