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

Public dataset limitations have significantly hindered the development andbenchmarking of learning to defer (L2D) algorithms, which aim to optimallycombine human and AI capabilities in hybrid decision-making systems. In suchsystems, human availability and domain-specific concerns introducedifficulties, while obtaining human predictions for training and evaluation iscostly. Financial fraud detection is a high-stakes setting where algorithms andhuman experts often work in tandem; however, there are no publicly availabledatasets for L2D concerning this important application of human-AI teaming. Tofill this gap in L2D research, we introduce the Financial Fraud Alert ReviewDataset (FiFAR), a synthetic bank account fraud detection dataset, containingthe predictions of a team of 50 highly complex and varied synthetic fraudanalysts, with varied bias and feature dependence. We also provide a realisticdefinition of human work capacity constraints, an aspect of L2D systems that isoften overlooked, allowing for extensive testing of assignment systems underreal-world conditions. We use our dataset to develop a capacity-aware L2Dmethod and rejection learning approach under realistic data availabilityconditions, and benchmark these baselines under an array of 300 distincttesting scenarios. We believe that this dataset will serve as a pivotalinstrument in facilitating a systematic, rigorous, reproducible, andtransparent evaluation and comparison of L2D methods, thereby fostering thedevelopment of more synergistic human-AI collaboration in decision-makingsystems. The public dataset and detailed synthetic expert information areavailable at: https://github.com/feedzai/fifar-dataset

FiFAR: A Fraud Detection Dataset for Learning to Defer