Title :
Multi-instance Learning for Predicting Fraudulent Financial Statements
Author :
Kotsiantis, Sotiris ; Kanellopoulos, Dimitris
Author_Institution :
Dept. of Comput. Sci. & Technol., Univ. of Peloponnese, Tripoli
Abstract :
This paper explores the effectiveness of multi-instance learning techniques in detecting firms that issue fraudulent financial statements (FFS). For this reason, a number of experiments have been conducted using representative learning algorithms, which were trained using a data set of 164 fraud and non-fraud Greek firms. The results show that MIBoost algorithm with decision stump as base learner had the best accuracy.
Keywords :
financial data processing; learning (artificial intelligence); MIBoost algorithm; decision stump; fraudulent financial statement prediction; multiinstance learning techniques; nonfraud Greek firms; representative learning algorithms; Audit Committee; Computer science; Europe; Financial management; Information technology; Mathematics; Quality management; Regulators; Security; Stock markets; classification; data mining; machine learning;
Conference_Titel :
Convergence and Hybrid Information Technology, 2008. ICCIT '08. Third International Conference on
Conference_Location :
Busan
Print_ISBN :
978-0-7695-3407-7
DOI :
10.1109/ICCIT.2008.150
