CRO-Guide-to-Machine-Learning

THE DATA

The following table compares the research findings to highlight which combination of models provides the highest prediction accuracy.

PREDICTION ACCURACY (%) FRAUD NON- FRAUD TOTAL

STUDY

METHOD/ TECHNIQUE

APPLICATION

Aleskerov et al (1997) Bell and Carcello (2000) Brause et al (1999)

ANNs

Credit Card

n/a

n/a

n/a

Financial Reporting

55

96

87

Statistical

Data mining and ANNs

Credit Card

n/a

n/a

n/a

Bolton and Hand (2002)

Clustering

Credit Card

n/a

n/a

n/a

Calderon and Green (1994) Dorronsoro et al (1997)

Financial Reporting

20

89

Statistical

n/a

ANNs

Credit Card

n/a

n/a

n/a

Ezawa and Norton (1996)

BNNs

Credit Card

n/a

n/a

n/a

Ghosh and Reilly (1994)

ANNs

Credit Card

n/a

n/a

n/a

Financial Reporting

68

74

72

Green and Choi (1997)

ANNs

Leonard (1995)

Expert System

Credit Card

n/a

n/a

n/a

Financial Reporting

35

86

76

Lin et al (2003)

FNNs

Quash and Sriganesh (2007) Zaslavsky and Strizkak (2006)

ANNs (Self-Organising Maps)

Credit Card

n/a

n/a

n/a

ANNs

Credit Card

n/a

n/a

n/a

Summary of the most notable investigations into the use of Artificial Intelligence at mitigating fraud.

The greatest challenge when talking about artificial intelligence/machine learning is actually in understanding what data sets we are looking at, and what model/combination of models to apply. Amazon’s Machine Learning offering is one example of an automated process which analyses the data and automatically selects the best model to use in the scenario. Other big players who have similar offerings are IBM, Google and Microsoft.

Made with FlippingBook Digital Publishing Software