Scribd
Upload
11 views2 pages

Main

The Biometric-Enhanced Transaction Monitoring Model uses Multi-Factor Authentication and biometric confirmation to secure real-time financial transactions. It processes biometric data to verify user identity and calculates a risk score that incorporates a biometric confidence score, enhancing transaction legitimacy assessment. The model also adapts over time to minimize false rejections while employing advanced encryption techniques to protect biometric data.

Uploaded by

nirmala periasamy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
11 views2 pages

Main

The Biometric-Enhanced Transaction Monitoring Model uses Multi-Factor Authentication and biometric confirmation to secure real-time financial transactions. It processes biometric data to verify user identity and calculates a risk score that incorporates a biometric confidence score, enhancing transaction legitimacy assessment. The model also adapts over time to minimize false rejections while employing advanced encryption techniques to protect biometric data.

Uploaded by

nirmala periasamy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Biometric-Enhanced

Transaction Monitoring
Model
Introduction
In this model, real-time financial transactions are monitored after users pass
through Multi-Factor Authentication (MFA) and provide biometric
confirmation. Biometric data, such as fingerprints or facial recognition, adds
an extra layer of security by uniquely verifying the user before transaction
approval.

Biometric Data Processing and


Verification
Once a user initiates a transaction, they provide biometric input (e.g., a
fingerprint). The biometric feature is processed to extract distinct data
points that represent unique characteristics. For example, a fingerprint scan
captures points such as ridges and valleys, creating a feature vector B of
biometric data:

B= { b1 ,b 2 , … , b n }

where each b i represents a specific feature point. The captured vector B is


compared against the stored biometric template T . A matching score M is
calculated as follows:
n
1
M= ∑ ∼( b i , t i)
n i=1
where t i represents points from the stored template, and ¿ ( b i , t i ) is a
similarity function measuring how closely the two points match. If M
exceeds a predefined threshold θ , the biometric input is accepted; otherwise,
the transaction is blocked.

Biometric-Enhanced Risk Scoring


Model
Upon successful biometric verification, transaction data is processed
further. The risk score R now includes a biometric confidence score S biometric ,
calculated as follows:

R=w1 × Samount + w2 × Sfrequency +w 3 × S location + w4 × S biometric


where w 1 , w 2 , w3, and w 4 are weights assigned to each factor based on
importance, and S biometric represents the confidence score derived from the
biometric match M . A higher biometric score lowers the overall risk,
indicating the transaction is likely legitimate.

Continuous Learning and


Adaptation
The model continuously learns from biometric patterns over time. For
example, if facial recognition slightly varies due to lighting, the model
updates the user's template within secure limits, minimizing false rejections
and maintaining security.

Security and Privacy


To protect biometric data, advanced encryption techniques (e.g., SHA-256
hashing) are applied during transmission and storage. Biometric templates
are stored in hashed, encrypted form to ensure data security.

You might also like