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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Credit card fraud detection is a very challenging problem because of thespecific nature of transaction data and the labeling process. The transactiondata is peculiar because they are obtained in a streaming fashion, they arestrongly imbalanced and prone to non-stationarity. The labeling is the outcomeof an active learning process, as every day human investigators contact only asmall number of cardholders (associated to the riskiest transactions) andobtain the class (fraud or genuine) of the related transactions. An adequateselection of the set of cardholders is therefore crucial for an efficient frauddetection process. In this paper, we present a number of active learningstrategies and we investigate their fraud detection accuracies. We comparedifferent criteria (supervised, semi-supervised and unsupervised) to queryunlabeled transactions. Finally, we highlight the existence of anexploitation/exploration trade-off for active learning in the context of frauddetection, which has so far been overlooked in the literature.

Streaming Active Learning Strategies for Real-Life Credit Card Fraud Detection: Assessment and Visualization