Solving the False Positives Problem in Fraud Prediction Using Automated Feature Engineering.

Wedge, R., Kanter, J. M., Veeramachaneni, K., Rubio, S. M., & Perez, S. I. (2018). Solving the False Positives Problem in Fraud Prediction Using Automated Feature Engineering. Machine Learning and Knowledge Discovery in Databases (ECML PKDD 2018).

Type

Book chapter

Intellectual base

Review or monograph

Year

2018

Volume

1105

Pages cited

372-388

DOI

10.1007/978-3-030-10997-4_23

ISBN

9783030109967

DOI registrant

Springer Nature

Link

https://doi.org/10.1007/978-3-030-10997-4_23

Cited from

systems_cybernetics

Cited by

1 citation across 1 artifact.

Each citation links to the sentence it supports in the citing article.

Primes

• Self Engagement Under Misclassification
  • Add classifier inputs for better discrimination — multi-spectral IFF, multi-receptor immune recognition, multi-feature fraud models.

    This sourceShows that adding classifier inputs (237 automatically engineered behavioral features) cut fraud false positives by ~54% while preserving detection — the classifier-side fix (more inputs, tuned to cost asymmetry) that reduces self-harm without weakening defense.

    Supported in partVerified against the work's full text

    Supports only the fraud-model example: many added behavioural features cut fraud false positives by 54%; the IFF and immune-recognition examples are outside this source.

    “We generate 237 features (>100 behavioral patterns) for each transaction, and use a random forest to learn a classifier.”

    Read by an automated reader; not a human review. How support was checked

Verification

Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.

Does it back the claim? Read against the text for 1 of 1 citation: 1 supported in part. Each verdict is shown under its citation below, with what in the work backs the sentence.

Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.

See how references were verified.

Registry ID ref:5ffd08aa8820 · see in the full table