A Silent Signature

An Introduction To Behavioral Biometrics

Sometimes we do a thing in order to find out the reason for it. Sometimes our actions are questions not answers — John LeCarre

Technology is eating the world one person at a time. There may be differences in the extent of digitization among countries but the broad direction is the same. For most technologies such as Financial Technology (FinTech) or the Blockchain, establishing a digital identity is critical to their implementation and long term success. As per Nuance Communications, Inc, fraud is a significant issue globally, and is growing fast in the contact center. In the US alone, more than 50% of the population has reported being a victim of some kind of fraud or data breach.

Today, a wide variety of authentication methods are in play. From Chip and PIN technology to biometrics, each authentication system has it’s own advantages and disadvantages. A better technology is taking shape. It is called behavioral biometrics and it combines non-invasive monitoring of user behavior and using that behavior along with biometrics to establish identity. This article provides an introduction to behavioral biometrics and to the ways in which it is changing the realm of digital identities and authentication systems.

Multi-Factor Authentication and Biometrics

The National Institute of Standards and Technology (NIST) provides organizations with a framework to deal with cyber security. About 30% of US companies adhered to these set of guidelines in 2015 with the number projected to increase to 50% in 2020. The NIST framework is based on five pillars i.e. identify, protect, detect, respond, recover. Multi-factor authentication is critical to achieving the requirements of this framework.

A factor is a piece of evidence that you are who you claim to be. For instance, a security question that asks you your best friends last name or the make of your first car is a piece of evidence that is ideally known only to you. Today, many authentication systems are multi-factor meaning they ask for two (two factor authentication) or more (multi-factor) pieces of evidence. These factors can be classified into three categories:

Inherence Factors: which include biometrics are are factors intrinsic to the user. Examples of authentication include fingerprint and retina scan.

Possession Factors: These factors are pieces of evidence generated by hardware or software devices called ‘tokens’. There are three broad categories of such tokens

Connected tokens (which are devices connected to the computer),
Disconnected tokens which are independent devices that display a number or alphanumeric code on a screen and,
Software tokens that are a two factor authentication device used to access computer services

Knowledge Factors: These factors deal with unique pieces of knowledge known only to the real user. Examples include date of birth, last four digits of social, city of birth etc.

Behavioral Biometrics

Behavioral biometrics is the field of study related to the measure of uniquely identifying and measurable patterns in human activities. The term contrasts with physical biometrics, which involves innate human characteristics such as fingerprints or iris patterns_(source:_ https://whatis.techtarget.com). This concept video from BehavioSec is a good visual demonstration of Behavioral Biometrics:

The biggest advantage of behavioral biometrics is that is it non-invasive. In other words, it does not obstruct user behavior like entering a password does. Let’s take an example, Sam presses keys on a keyboard in a particular manner. He also uses the track pad in a particular manner. Software is used to capture these two behavioral data points. Next time, the company collecting these data points (lets say a bank) wants to authenticate Sam, it use machine learning algorithms to match Sam’s behavioral data points to the his actual keystrokes and track pad movement to identify Sam without creating friction during his user experience.

source: paymentscardsandmobile.com

Many Here is an example of IBM using Behavioral Biometrics to prevent car theft.

Another example is Bio Catch which is helping financial institutions verify identity using behavior. As per BioCatch, it selects 20 unique features from its 500+ patented behavioral profiling metrics to analyze a user’s behavior throughout a session, without any disruption in the digital experience. BioCatch compares the user behavior in real-time against the profile to return an actionable risk score.

Nuance communications is one of the pioneers in the field of voice authentication. It was the first company to release voice biometric algorithms powered by deep neural networks in 2015. Nuance offers a solution called ‘ConversationPrint’ which can authenticate users in real time using patterns of speech or patterns in writing.

Thus, there are a myriad of applications that can benefit from the proliferation of Behavioral Biometrics.

The Human Side Of Technology

The United Nations World Food Program (WFP) recently launched a pilot program in Jordan and Pakistan called Building Blocks. It launched a proof of concept in the Sindh province of Pakistan in 2017 and later launched a pilot in Jordan which leveraged it’s learning in Pakistan. As per it’s website, WFP is testing Blockchain as a means of making cash transfers more efficient, transparent and secure. This program is unique because it uses biometric authentication to authenticate the recipient of humanitarian aid, queries an account recorded on the Blockchain and allows the recipient to settle the payment. In April 2018, The MIT Technology Review covered this fascinating story in detail.

Now, imagine if Behavioral Biometrics could be used to identify recipients and then disburse charity or money transfers using data recorded on the Blockchain. Such applications would change the world of giving forever by eliminating the middleman. Behavioral Biometrics also has great potential to spread financial inclusion to the masses by helping establish trust.

My father is a financial adviser. Every morning, he would wake up and look at the scrolling ticker while jotting down the stock prices on his notepad. Quite naturally, I would spend most of my mornings watching CNBC. During my summer breaks, I would accompany him to the Bombay Stock Exchange and I would meet brokers and other market participants. Suffice it to say that finance has been my favorite subject since I was a teenage kid who thought nothing was impossible.

I still think anything is possible. I want my readers to rediscover the potential of technology to do good. As always, I wanted to bring the behavioral biometrics story home to the people it should serve. In doing so, I wanted to inject positivity in a fascinating technological development. Especially, when positivity is in short supply.

A Silent Signature

An Introduction To Behavioral Biometrics

Multi-Factor Authentication and Biometrics

Behavioral Biometrics

The Human Side Of Technology

Resources: