Under the appearance of combating financial crime, behavioral biometry is developed, which, thanks to the collection of many data on user interaction with devices, provides for human behaviour – allegedly only to fight possible fraudsters.
TransUnion estimates that the costs of scamming companies are now almost 7 percent of their revenue. American companies were to lose out due to this kind of fraud a full of about $1 112 billion last year.
To defend themselves from losses, companies are to adopt more advanced technologies to safe their operations. Artificial Intelligence (AI) is the key.
Algorithmic software shall be integrated into the platforms utilized to counter money laundering (AML). For example, Napier AI and ThreatMark have developed a fraud detection package to improve compliance and increase protection against illegal financial activities. The cooperation uses AI to analyse real-time data to detect anomalies that may indicate possibly fraudulent behaviour.
Another Experian UK&I company established cooperation with Sardine, startup preventing fraud utilizing behavioral biometry and device intelligence technology.
Behavioral biometry and continuous detection technologies are gaining popularity. BioCatch introduced a rules simulator to aid test and optimize real-time fraud detection strategies.
Globally, there are intelligence platforms focusing on detecting and preventing fraud. For example, Trulioo has late launched a global " intelligence capacity" based on a network of data points to identify fraud patterns in different countries. The hazard indicators shall be combined with the consortium data by monitoring data transmitted in real time through a single API. Trulioo fraud intelligence is to be launched in 195 countries, for sectors specified as payments, banking and markets.
Behavioral biometry is an advanced way of confirming the identity of people by examining their behaviour online. For this purpose, algorithmic solutions and alleged device learning are utilized to transform behaviour into biometric data.
This technology goes beyond conventional data sets and learns how users physically and behaviorally interact with their devices. Physical biometric data, specified as fingerprints and retinal scans, are based on unchanged features. Behavioral biometry is distinguished by the survey of unique individual actions, e.g. analyses the way the keyboard is written, the rate of finger movement on the contact screen or mouse movements. These activities contain large amounts of data to separate users.
Banks analyse behaviour online, trying to detect an anomaly and possible fraud in real time. For example, a unique pattern of user writing velocity or keystroke moves creates the basis for creating a matrix of average writing behavior. abrupt changes in writing velocity or pressing keys can trigger alerts due to possible fraud.
Individual mouse movements and interactions with the website supply additional biometric data. Like the way you contact and decision content on the device screen.
Behavioral biometry is thus different from fraud-proof solutions, which include identity checking erstwhile logging in that it allows you to analyse your behaviour throughout a payment service session.
Unfortunately, it frequently causes a number of problems for customers due to false alarms.
Advanced behavioral biometry techniques are designed to aid and detect bot activities, including sophisticated precursor attacks, offering a proactive approach to fraud prevention.
Biometric behavioral data offers a deep insight into the habits, trends and preferences of the individual, which may consequence in discrimination. The problem is data breach by stealing from hackers who can impersonate legitimate clients and gain access to their accounts.
What's being analyzed? For example, the dynamics of writing is measured by rhythm, velocity and way of writing by a individual (e.g. the frequency of utilizing the left or right button). Shift, mistake rate, etc.).
The technology measures the force on the keys, the waiting time (the time of pressing the key) and the flight time (the time between releasing the key and pressing the next key). This indicates the rhythm and kind of writing. It monitors the unique pressing of user keys during the session and compares them with historical data to look for different behaviors.
Similarly, behavioral biometric systems can track user interaction with a computer mouse or Touchpad, including the moves and clicks he makes. Basically, the strategy will track interactions specified as mouse location, the dimension and strength of buttons clicks, the velocity of mouse movement and others. Even tiny hand movements and gestures are detected and monitored utilizing behavioral biometric technology.
Each user has a unique pattern and kind of utilizing the mouse, which helps systems find erstwhile the user of the device is an authorized account holder and erstwhile suspicious activity occurs.
User interaction with the contact screen can besides be utilized to monitor suspicious activity on the device. The way it scrolls up or down, the force it exerts on the screen, and the velocity of interaction are crucial data that aid separate users from themselves and human users from bots.
In future, all actions on the contact screen will be monitored and compared with stored data.
The way a individual deals with his device is another crucial behavioral biometric indicator. 2 sensors: the gyroscope measures the rotation and orientation of the device, and the accelerator reflects the acceleration of the movement of the device. You can so make profiles of typical movements and user behaviour.
While physical biometry focuses on the innate unique physical characteristics of a person, e.g. fingerprints, iris/retina structure, facial features, vein pattern, voice pattern, behavioral biometry focuses on acquired behaviour patterns. In another words, it focuses on what the user does, not on who he is.
If safety is breached and the digitalised physical biometric data of users is revealed, this could pose a serious hazard in the long term. In the event of a breach of the password, they can be changed, updated, etc. Biometric data can't be changed. They enable continuous analysis of current user interactions in a strategy or application, but this technology is not accurate and can make false affirmative or negative results that affect environmental factors, natural variability of behaviour and complexity of any of their patterns.
The user has the right to feel uncomfortable knowing that his activity is constantly monitored, and behaviour patterns are stored and analyzed.
This kind of operation besides requires devices with advanced computing power and memory, which generates a number of social problems associated with the creation of fresh data centers that consume immense amounts of energy, water, etc.
Digitalisation in any areas progresses rapidly. The EU authorities have late proposed that both EU citizens and third-country nationals are subject to systematic individual checks erstwhile crossing the EU's external borders, the usage of digital passports and identity cards. The European Commission wants a common framework for the usage of digital identity papers through the EU Digital Travel application.
Digital IDs are to be stored on a mobile telephone and in the EU Digital Travel application.
The EU authorities point out that the usage of digital identity cards "will improve security: fraudsters will be more hard to usage false papers or to cross Schengen borders unnoticed".
The fresh solution is to improve the effectiveness of border controls, let for smoother and faster border crossings (travels will be able to present digital passports or IDs for prior border control before travel). This is besides intended to reduce the administrative burden on EU citizens.
The application will aid analyse the behaviour of travelers based on previously sent travel plans and papers to border services in advance.
Sources: biometricupdate.com, feedzai.com, commission.europa.eu
AS
