Assessing Methodologies for Intelligent Bankruptcy Prediction

Kirkos, Efstathios/ Κύρκος, Ευστάθιος

Institution and School/Department of submitter: ΤΕΙ Θεσσαλονίκης
Keywords: Intelligent techniques;Credit scoring;Business intelligence;Bankruptcy prediction
Issue Date: 8-Nov-2012
Publisher: Springer Netherlands
Citation: Kirkos, E. (8 Οκτωβρίου 2012). Assessing methodologies for intelligent bankruptcy prediction. Artificial intelligence review 43, (1). Διαθέσιμο σε: (Ανακτήθηκε 30 Ιουνίου 2015).
Journal: Artificial Intelligence Review, vol. 43, no. 1, 2012
Abstract: Bankruptcy prediction is one of the most important business decision-making problems. Intelligent techniques have been employed in order to develop models capable of predicting business failure cases. The present article provides a systematic literature review of the field. As opposed to previous reviews which concentrate on the classification methods, this study adopts a much broader approach to the bankruptcy prediction problem. The survey is articulated around six major axes which cover all the range of issues related to bankruptcy prediction. These axes are the definition of main research objectives, the employed classification methods, performance metrics issues, the input data and data sets, feature selection and input vectors and finally, the interpretation of the models and the extraction of domain knowledge. The findings and employed methodologies of the collected papers are categorized, presented and assessed according to these axes. The ultimate goal is to detect weaknesses and omissions and to highlight research opportunities. We hope that future researchers will find this survey useful in their attempt to orientate their efforts and to locate interesting topics for further research.
Description: Δημοσιεύσεις μελών--ΣΔΟ--Τμήμα Λογιστικής, 2012
ISSN: 0269-2821
Other Identifiers: 10.1007/s10462-012-9367-6
Appears in Collections:Δημοσιεύσεις σε Περιοδικά

Files in This Item:
There are no files associated with this item.

 Please use this identifier to cite or link to this item:
  This item is a favorite for 0 people.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.