Title: A modified fruit fly optimisation for classification of financial distress using FLANN
Authors: Soumya Ranjan Sahu; Devi Prasanna Kanungo; Himansu Sekhar Behera
Addresses: Department of Computer Science and Engineering and Information Technology, Veer Surendra Sai University of Technology (VSSUT), Burla, Sambalpur 768018, Odisha, India ' Department of Computer Science and Engineering and Information Technology, Veer Surendra Sai University of Technology (VSSUT), Burla, Sambalpur 768018, Odisha, India ' Department of Computer Science and Engineering and Information Technology, Veer Surendra Sai University of Technology (VSSUT), Burla, Sambalpur 768018, Odisha, India
Abstract: Current financial market has become proficient enough to cater to the needs of a large customer base. But, at the same time, the number of market catastrophes is on the rise leading to the increase in the demand of precise and potential classifier models. In this work, a hybrid model of clustering and neural network-based classifier has been proposed, i.e., FCM-FLANN-IFFO. Three financial credit risk datasets were applied in the experiment and the model is evaluated using 12 different performance metrics. This novel metaheuristic uses an improved version of fruit fly algorithm which is inspired by the foraging behaviour of the fruit flies to locate their food. The experimental results illustrates that proposed model outperforms other models. The proposed model provides brilliant results with 94.91% of classification accuracy.
Keywords: classification; FLANN; functional link artificial neural network; financial credit risk; CRO; chemical reaction optimisation; clustering; FCM; fuzzy C-means; FOA; fruit fly optimisation; IFFO; improved fruit fly optimisation.
International Journal of Data Science, 2017 Vol.2 No.2, pp.155 - 172
Received: 14 May 2015
Accepted: 02 Feb 2016
Published online: 26 Jun 2017 *