loader image
Close
  • Use Case
    • Banking and finance
    • E-commerce and retail
    • Telecommunications and media providers
    • Administration
    • Healthcare
    • Technology and IT
  • Types of activity
    • Account sharing and device sharing
    • Employee authentication
    • Hardware-free MFA/ 2FA for clients
    • Remote working
    • Compliance with PSD2, SCA regulations
    • Anti-fraud system
    • Detection of bots and remote desktops
  • Solution
    • Behavioural biometrics
    • Device Fingerprinting
    • PureSecure
    • MobileSecure
  • Knowledge zone
    • Onepgers
    • Presentations
    • E-book
    • Webinars
    • Video
  • News
    • Blog
    • Expert texts
  • Company
    • About us
    • About BIK group
    • Privacy policy
    • European union
  • Contact
  • Use Case

    Banking and finance

    E-commerce and retail

    Telecommunications and media providers

    Administration

    Healthcare

    Technology and IT

    Account sharing and device sharing

    Employee authentication

    Hardware-free MFA/ 2FA for clients

    Remote working

    Compliance with PSD2, SCA regulations

    Anti-fraud system

    Detection of bots and remote desktops

  • Solution

    Behavioural biometrics

    Device Fingerprinting

    PureSecure

    MobileSecure

  • Knowledge zone

    Onepgers

    Presentation

    E-book

    Webinars

    Video

  • News

    Blog

    Expert texts

  • Company

    About us

    About BIK group

    Privacy policy

    European Union Projects

  • Contact
  • English
    • Polski
    • English

  • Use Case

    Banking and finance

    E-commerce and retail

    Telecommunications and media providers

    Administration

    Healthcare

    Technology and IT

    Account sharing and device sharing

    Employee authentication

    Hardware-free MFA/ 2FA for clients

    Remote working

    Compliance with PSD2, SCA regulations

    Anti-fraud system

    Detection of bots and remote desktops

  • Solution

    Behavioural biometrics

    Device Fingerprinting

    PureSecure

    MobileSecure

  • Knowledge zone

    Onepgers

    Presentation

    E-book

    Webinars

    Video

  • News

    Blog

    Expert texts

  • Company

    About us

    About BIK group

    Privacy policy

    European Union Projects

  • Contact
  • English
    • Polski
    • English

Knowledge

Models’ quality adapted to your needs

For each user, Digital Fingerprints can provide independent models based on users’ behaviour – one runs on a keyboard, another on mouse/touchpad etc. activity. But what does it really mean?

We get some data about how the user behaves – and then we need to make a decision – whether it was the real user or not?

If you know a bit about statistics you are surely aware that there is no way to be free from some kind of error. In binary classification – like ours – we can have two types of errors:

  1. We predict someone else on account when it was actually a real user, or
  2. We will miss fraud because we thought it was user while it was someone else using his/her account

The most interesting thing – both for us and for our clients – would be to determine which of these two are being made.

Very popular and easy-to-understand way to see what is the accuracy of our classifier is a confusion matrix. Let me show you a simplified version of it, which shows how it is done in Digital Fingerprints:

In each field, we would put a number of cases when we classified user as a user etc. And then we can count many kinds of metrics which will show us how good we are from different points of view. Let’s start from the most popular one – accuracy (ACC) – which overall answers the question: how often are we correct?

The closer to 1 accuracy is, the fewer mistakes our classifier is making.

Other important metrics are sensitivity (SE) and specificity (SP).

The plot displays an example of how those two metrics can depend on a threshold. As you can see – both lines go quite similar – that’s why more useful would be to see ROC curve – but first, let’s get to know two more metrics.

If we want to focus more on mistakes we can validate false negative rate (FNR) and false positive rate (FPR). Those metrics are very helpful while making a decision which threshold is best for each user.

 

False Positive Rate is actually equal to 1 – Specificity. If we want to combine all those results into one plot – here ROC curve comes into play. Below are some examples of how it may look:

AUC – the Area Under Curve gives us the idea of how a model is performing. To make the long story short – the larger the area is, the better model we have. So this simply indicates how well the probabilities from the positive classes are separated from the negative ones.

According to those (and actually many others) metrics we can choose the best possible model accuracy which will fit your needs.

Just tell us about your expectations and we will figure out the best solution!

References

  1. James, G., Witten, D.., Hastie, T., Tibshirani, R. (2013). An Introduction to Statistical Learning with Applications in R. Springer.
  2. Hastie, T., Tibshirani, R., & Friedman, J. H. (2001). The elements of statistical learning: Data mining, inference, and prediction. Springer.

Several million people in Poland will test Behavioural Biometrics - interview in Subiektywnieofinansach.pl
Previous Article
Behavioural Biometry and SCA requirements
Next Article

Digital Fingerprints © Copyright 2022
Created by <code-One>

Use Case

Banking and finance
E-commerce and retail
Telecommunications and media providers
Administration
Healthcare
Technology and IT

Solutions

Behavioural biometrics
Device Fingerprinting
PureSecure
MobileSecure
About us
About group
EU project
Privacy policy
Contact

Digital Fingerprints S.A. ul. Żeliwna 38, 40-599 Katowice. KRS: 0000543443, Sąd Rejonowy Katowice-Wschód, VIII Wydział Gospodarczy, Kapitał zakładowy: 128 828,76 zł – opłacony w całości, NIP: 525-260-93-29

Biuro Informacji Kredytowej S.A., ul. Zygmunta Modzelewskiego 77a, 02-679 Warszawa. Numer KRS: 0000110015, Sąd Rejonowy m.st. Warszawy, XIII Wydział Gospodarczy, kapitał zakładowy 15.550.000 zł opłacony w całości, NIP: 951-177-86-33, REGON: 012845863.

Biuro Informacji Gospodarczej InfoMonitor S.A., ul. Zygmunta Modzelewskiego 77a, 02-679 Warszawa. Numer KRS: 0000201192, Sąd Rejonowy m.st. Warszawy, XIII Wydział Gospodarczy, kapitał zakładowy 7.105.000 zł opłacony w całości, NIP: 526-274-43-07, REGON: 015625240.

Zarządzaj zgodami plików cookie
Aby zapewnić jak najlepsze wrażenia, korzystamy z technologii, takich jak pliki cookie, do przechowywania i/lub uzyskiwania dostępu do informacji o urządzeniu. Zgoda na te technologie pozwoli nam przetwarzać dane, takie jak zachowanie podczas przeglądania lub unikalne identyfikatory na tej stronie. Brak wyrażenia zgody lub wycofanie zgody może niekorzystnie wpłynąć na niektóre cechy i funkcje.
Funkcjonalne Always active
Przechowywanie lub dostęp do danych technicznych jest ściśle konieczny do uzasadnionego celu umożliwienia korzystania z konkretnej usługi wyraźnie żądanej przez subskrybenta lub użytkownika, lub wyłącznie w celu przeprowadzenia transmisji komunikatu przez sieć łączności elektronicznej.
Preferences
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
Statystyka
Przechowywanie techniczne lub dostęp, który jest używany wyłącznie do celów statystycznych. The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
Marketing
Przechowywanie lub dostęp techniczny jest wymagany do tworzenia profili użytkowników w celu wysyłania reklam lub śledzenia użytkownika na stronie internetowej lub na kilku stronach internetowych w podobnych celach marketingowych.
Manage options Manage services Manage vendors Read more about these purposes
Zobacz preferencje
{title} {title} {title}