Developers: | ID Finance |
Date of the premiere of the system: | 2017 |
Technology: | Cybersecurity - Biometric identification, cybersecurity - the Fraud detection system (fraud) |
The ID Finance financial technical-company provided the system of behavioural biometrics capable to determine the person by keyboard handwriting in 2018. The pilot version was implemented in the Russian MoneyMan project (enters into ID Finance) in August, 2017.
A system allows to carry out user authentication on the behavioural patterns inherent to each person. Among them: speed and dynamics of text typing, transition time between keys, typos and features of the movement of the cursor of a mouse on the screen. As soon as the person is registered in a personal account, a system analyze his behavior and does a mold which is used at the subsequent authentication on the website, the director of ID Finance risk management the Cossack Ekaterina noted.
The program excludes fraud cases on the website. If at logging into the personal account the handwriting of the user does not match a mold in base, such profile is entered in the list risky, requiring additional check.
"If there are serious reasons to assume that the person not that for whom issues itself we can request optional data, including the photo of the passport. Behavioural patterns cannot be forged: even if the person replaced the keyboard they can slightly change, but in total their structure will remain invariable", - told the Cossack.
A system was developed by own forces of holding on the basis of technology of the analysis of Big Data. On development and deployment of the program about 4 weeks left, the pilot lasted 6 months. Accuracy of the pilot version of a biometric system was 97.6%. It is improved, accuracy will increase. A biometric system formed the basis a fraud scoring of ID Finance which allows to identify swindlers. The biometrics is used not only at authentication, but also at primary registration when the handwriting of the client is compared to all molds in base.
Thanks to a system the antifraud scoring was succeeded to reduce the cost of one issued loan almost by 25% due to cutting of costs for verification and costs of requests in information sources. Level of approval of requests at the same time grew by 28%, follows from results of work for half a year. Economic effect of implementation is estimated more than at 90 million rubles in Russia at the end of 2018. Economic effect only of a sensing technology of keyboard handwriting on the ID Finance group in seven countries is estimated at $2.8 million at the end of 2018.