Multiobjective Evolutionary Optimization of Type-2 Fuzzy Rule-Based Systems for Financial Data Classification

Abstract

Abstract-Classification techniques are becoming essential in the financial world for reducing risks and possible disasters. Managers are interested in not only high accuracy but also in interpretability and transparency. It is widely accepted now that the comprehension of how inputs and output are related to each other is crucial for taking operative and strategic decisions. Furthermore, inputs are often affected by contextual factors and characterized by a high level of uncertainty. In addition, financial data are usually highly skewed towards the majority class. With the aim of achieving high accuracies, preserving the interpretability and managing uncertain and unbalanced data, the paper presents a novel method to deal with financial data classification by adopting type-2 fuzzy rule-based classifiers (FRBCs) generated from data by a multi-objective evolutionary algorithm (MOEA). The classifiers employ an approach, denoted as scaled dominance, for defining rule weights in such a way to help minority classes to be correctly classified. In particular, we have extended PAES-RCS, an MOEA-based approach to learn concurrently the rule and data bases of FRBCs, for managing both interval type-2 fuzzy sets and unbalanced datasets. To the best of our knowledge, this is the first work that generates type-2 FRBCs by concurrently maximizing accuracy and minimizing the number of rules and the rule length with the objective of producing interpretable models of real-world skewed and incomplete financial datasets. The rule bases are generated by exploiting a rule and condition selection (RCS) approach, which selects a reduced number of rules from a heuristically generated rule base and a reduced number of conditions for each selected rule during the evolutionary process. The weight associated with each rule is scaled by the scaled dominance approach on the fuzzy frequency of the output class, in order to give a higher weight to the minority class. As regards the data base learning, the membership function parameters of the interval type-2 fuzzy sets used in the rules are learned concurrently to the application of RCS. Unbalanced datasets are managed by using, in addition to complexity, selectivity and specificity as objectives of the MOEA rather than only the classification rate. We tested our approach, named IT2-PAES-RCS, on eleven financial datasets and compared our results with the ones obtained by the original PAES-RCS with three objectives and with and without scaled dominance, the fuzzy rule-based classifiers FARC-HD and FURIA, the classical C4.5 decision tree algorithm and its costsensitive version. Using non-parametric statistical tests, we will show that IT2-PAES-RCS generates FRBCs with, on average, accuracy statistically comparable to and complexity lower than the ones generated by the two versions of the original PAES-RCS. Further, the FRBCs generated by FARC-HD and FURIA and the decision trees computed by C4.5 and its cost-sensitive version, despite the highest complexity, result to be less accurate than the FRBCs...

CFM-BD: A Distributed Rule Induction Algorithm for Building Compact Fuzzy Models in Big Data Classification Problems

“…Humberto Bustince is also with King Abdulaziz University, Jedda, Saudi Arabia. E-mails: {mikel.elkano, joseantonio.sanz, edurne.barrenechea, bustince, mikel.galar}@unavarra.es fault prediction [4], anomaly intrusion detection [5], financial applications [6], image processing [7], and traffic congestion prediction [8], among others.…”

Toward Human-Understandable, Explainable AI

“…Furthermore, FRBSs werenot able to handle easily imbalanced and skewed data (such as those present in fraud, bank default data, etc). However, recent work such as [18], [19] was able to use evolutionary systems to generate FRBSs with short IF-Then rules and small number of rules in the rule base while maximizing the prediction accuracy. As this created sparse rule base not covering the whole search space, they presented a similarity technique to classify the incoming examples even if they do not match any fuzzy rule in the generated rule base.…”

“…antecedents even if the system had thousands of inputs as well as having a small rule base) and maximize the accuracy of the FLS prediction. It was shown in [18], [19] that such highly interpretable systems outperform decision trees like C4.5 by a big margin in accuracy while…”

Multiobjective evolutionary feature selection and fuzzy classification of contact centre data

“…In Ducange, Mannara, Marcelloni, Pecori, and Vecchio (), PAES‐RCS method is used to maximize accuracy and minimize the total rule length for internet traffic classification. In Antonelli, Bernardo, Hagras, and Marcelloni (), IT2‐PAES‐RCS extends PAES‐RCS to employ Type‐2 fuzzy sets, where sensitivity, specificity, and total rule length are optimized for financial data classification. Finally, a distributed version of PAES‐RCS by using Apache Spark as data processing framework is proposed in Ferranti, Marcelloni, and Segatori (), and in Ferranti, Marcelloni, Segatori, Antonelli, and Ducange ().…”