
IBM recently announced the acquisition of a patent for a new technology to combat online fraud, based on monitoring and analysis of behavioral factors. The technology itself is called “Fraud detection system based on analysis of user and browser interactions”.
Simply put, the system analyzes the user's behavior and his method of working on the Web, with certain sites, including banking resources and online shopping. Any user has a certain "line of conduct" on such sites, clicking on certain interface elements a certain number of times, using more keyboard shortcuts or a mouse. In addition, a particular user works with a particular device, and rarely changes the line of their behavior.
')
But everything changes, if the control intercepts the bot / attacker-man, interaction with the browser, a kind of "fingerprints" immediately changes. In such a case, the system developed by IBM begins to require additional identification methods from the user.
Of course, it may happen that the user's behavior changes due to a hand injury, for example, or a change in the working device (desktop PC to ultrabook with touchscreen or tablet). In this case, the person does not cost anything to go through additional identification.
But, according to the developers, the abrupt changes in behavioral factors in the vast majority of cases are caused by the actions of Internet fraudsters. And it will be much more difficult for attackers to go through additional identification than to the conditional user whose data was stolen.
The patent itself can be viewed here:
US Patent # 8,650,080