Machine Learning Detection for Financial Statement Fraud

Ting Kai Hwang*, Wei Chun Chen, Wan Chi Chiang, Yung Ming Li

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

This study intends to develop a methodology of fraudulent transaction detection model. The algorithm of XGBoost integrating the techniques of SMOTE sampling method and Bayesian Hyperparameter Optimization, is proposed to separate fraud transactions from non-fraud transactions. The experimental results based on the public data set of financial statement fraud from Kaggle website show the proposed model is better than the commonly used binary-classification methods, such as Logistic Regression, SVM, KNN, Random Forest, XGBoost without Hyperparameter Tuning and Multilayer Perceptron. The method of establishing fraud detection models assists people who lack the machine learning domain expertise for the modeling and tuning parameter techniques. It can help to detect abnormal transactions as early as possible and carry out risk management for banking industry.

Original languageEnglish
Title of host publicationInformation Systems and Technologies - WorldCIST 2022
EditorsAlvaro Rocha, Hojjat Adeli, Gintautas Dzemyda, Fernando Moreira
PublisherSpringer Science and Business Media Deutschland GmbH
Pages148-154
Number of pages7
ISBN (Print)9783031048180
DOIs
StatePublished - 2022
Event10th World Conference on Information Systems and Technologies, WorldCIST 2022 - Budva, Montenegro
Duration: 12 Apr 202214 Apr 2022

Publication series

NameLecture Notes in Networks and Systems
Volume469 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

Conference10th World Conference on Information Systems and Technologies, WorldCIST 2022
Country/TerritoryMontenegro
CityBudva
Period12/04/2214/04/22

Keywords

  • Bayesian optimization
  • Financial statement fraud
  • Hyperparameter tuning technologies

Fingerprint

Dive into the research topics of 'Machine Learning Detection for Financial Statement Fraud'. Together they form a unique fingerprint.

Cite this