English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The growing sophistication of Business Email Compromise (BEC) and spearphishing attacks poses significant challenges to organizations worldwide. Thetechniques featured in traditional spam and phishing detection are insufficientdue to the tailored nature of modern BEC attacks as they often blend in withthe regular benign traffic. Recent advances in machine learning, particularlyin Natural Language Understanding (NLU), offer a promising avenue for combatingsuch attacks but in a practical system, due to limitations such as dataavailability, operational costs, verdict explainability requirements or a needto robustly evolve the system, it is essential to combine multiple approachestogether. We present CAPE, a comprehensive and efficient system for BECdetection that has been proven in a production environment for a period of overtwo years. Rather than being a single model, CAPE is a system that combinesindependent ML models and algorithms detecting BEC-related behaviors acrossvarious email modalities such as text, images, metadata and the email'scommunication context. This decomposition makes CAPE's verdicts naturallyexplainable. In the paper, we describe the design principles and constraintsbehind its architecture, as well as the challenges of model design, evaluationand adapting the system continuously through a Bayesian approach that combineslimited data with domain knowledge. Furthermore, we elaborate on severalspecific behavioral detectors, such as those based on Transformer neuralarchitectures.

A Modular and Adaptive System for Business Email Compromise Detection