Artificial Bee Colony for Logic Mining in Credit Scoring
Open Access
- 29 December 2022
- journal article
- research article
- Published by Penerbit UTM Press in Malaysian Journal of Fundamental and Applied Sciences
- Vol. 18 (6) , 654-673
Abstract
During the SARS-CoV-2 (Covid-19) pandemic, credit applications skyrocketed unimaginably. Thus, creditors or financial entities were burdened with information overload to ensure they provided the proper credit to the right person. The existing methods employed by financial entities were prone to overfitting and did not provide any information regarding the behavior of the creditor. However, the outcome did not consider the attribute of the creditor that led to the default outcome. In this paper, a swarm intelligence-based algorithm named Artificial Bee Colony has been implemented to optimize the learning phase of the Hopfield Neural Network with 2 Satisfiability-based Reverse Analysis Methods. The proposed hybrid model will be used to extract logical information in the credit data with more than 80% accuracy compared to the existing method. The effectiveness of the proposed hybrid model was evaluated and showed superior results compared to other models.Keywords
This publication has 25 references indexed in Scilit:
- PRO2SAT: Systematic Probabilistic Satisfiability logic in Discrete Hopfield Neural NetworkAdvances in Engineering Software, 2022
- Novel logic mining incorporating log linear approachJournal of King Saud University Computer and Information Sciences, 2022
- An Automatic Credit Scoring Strategy (ACSS) using memetic evolutionary algorithm and neural architecture searchApplied Soft Computing, 2021
- A machine learning approach combining expert knowledge with genetic algorithms in feature selection for credit risk assessmentApplied Soft Computing, 2021
- A multi-objective approach for profit-driven feature selection in credit scoringDecision Support Systems, 2019
- Binary Artificial Bee Colony optimization using bitwise operationComputers & Industrial Engineering, 2014
- Logic mining in neural network: reverse analysis methodComputing, 2010
- The comparisons of data mining techniques for the predictive accuracy of probability of default of credit card clientsExpert Systems with Applications, 2009
- “Neural” computation of decisions in optimization problemsBiological Cybernetics, 1985
- Linear-time algorithms for testing the satisfiability of propositional horn formulaeThe Journal of Logic Programming, 1984