Fake Online Profiles Now Easier to Identify

Computer scientists from the University of Edinburgh have developed a new tool that can identify fake online profiles with relative ease.
The new system consists of computer models that were trained to spot online users who make up information about themselves – catfishes. The model can identify social media users who are dishonest about their gender and age and could, therefore, be useful in ensuring the safety of social networks.
The team built the new tool based on information they gathered from about 5,000 verified public profiles on an adult content website. These verified profiles were used to train the model to estimate the gender and age of a user with high accuracy. The system was able to estimate this based on the users’ style of writing in comments, as well as on their network activity.
Interestingly, the system found that almost 40% of the site’s users lie about their age, and about one-quarter lie about their gender.
The new tool demonstrates the effectiveness of the technology in fishing out dishonest users. The team hopes that this system will lead to safer and more honest world of social networks.

Source:
University of Edinburgh via ScienceDaily (https://www.sciencedaily.com/releases/2017/06/170607123855.htm)