Fuzzy logic system for identity theft detection in social networks

Abstract

Social networks are very popular in today’s world and have great importance for the interconnection between people. That is why the theft of access credentials to enter them is the order of the day. Many users are attacked by cybercriminals in various ways to seize their users and passwords, as both are a gateway to get information from the users, claim ransoms for their accounts or perform spamming techniques such as phishing to their contacts. In order to detect identity theft, this paper proposes the use of a text-mining and fuzzy logic-based system to decide if a person is being victimized by this problem. To do so, the natural language of each user is taken into account, considering values such as the total frequency of appearance of words and expressions they do not frequently use. Other factors such as the location from which users log on or the number of messages sent in a short period of time are also taken into account to make the final decision.