文章基本信息

标题：Bankruptcy Prediction using Hybrid Neural Networks with Artificial Bee Colony
本地全文：下载
作者：Said Marso ; Mohamed EL Merouani
期刊名称：Engineering Letters
印刷版ISSN：1816-093X
电子版ISSN：1816-0948
出版年度：2020
卷号：28
期号：4
页码：1191-1200
语种：English
出版社：Newswood Ltd
摘要：Credit risk now is considered like the most important risk faced by banks and financial institutions. For this reason, predictive analytics be- came the barometer of financiers with a main in- terest to corporate bankruptcy prediction (CBP), or referred to as financial distress prediction. Predictive analytics techniques are subdivided into two groups: traditional statistical (e.g. discriminant analysis and logistic regression) and machine learning (e.g. arti- ficial neural networks, support vector machines and random forest). Trained by metaheuristic algorithm such as artificial bee colony (ABC), an ANN model called ABCNN is applied in CBP and this new hy- brid model is the contribution of our article. Our model is compared with two other models in order to investigate its efficiency: the first model is multiple discriminant analysis (MDA) and the second one is an ANN trained by the most common learning algorithm a back propagation (BPNN).
关键词：Corporate bankruptcy prediction; artificial neural networks; artificial bee colony; train- ing; imbalanced data; machine learning.