How to identify suspicious aspects of online behavior

This article follows the previous one about fighting the fact that 10% of us is demonstrably lying while filling in an online form.

After a user begins to type in an online form you can track all possible aspects of his behavior from the keystroke dynamics through mouse movements to the time spent by filling in one or more fields of the form. This creates a unique biometric profile for each visitor. Thanks to it we are able to find possible suspicious aspects of his behavior. No matter if it is the frequent use of “copypaste” or even auto-filling, frequent returning to previous steps, rewriting of particular fields, change of personal data and many other actions.

Each of these metrics has a certain weight that affects the total score of each visitor - for example, the score of fraud probability. If the score is higher than the set threshold, tools like Zoe.ai can trigger a real-time alert of this event and handover it in the company's environment for deeper investigation. Thanks to our advanced dashboard you can easily filter between individual requests and their suspicious aspects.

Although we mainly deal with real-time online fraud in insurance companies, thanks to the behavioral detection of each user Zoe.ai is also ready to predict potential future fraud at the time of insurance arrangement or to estimate a price sensitivity of a particular user and thus help with customer segmentation and dynamic pricing.

Are you more interested to learn more? Feel free to write me!