Interpretability of linguistic fuzzy rule-based systems: An overview of interpretability measures

Multicriteria Design of Cost-Conscious Fuzzy Rule-Based Classifiers

Many real-world classification systems must comply with a series of inherent restrictions to the problem at hand such as response times, power consumptions or computational costs. This poses a fundamental limitation to traditional performance-driven classifiers and learning algorithms by restraining their applicability in cost-sensitive scenarios. Because of this, fuzzy systems are leveraged to learn cost-conscious multi-stage classifiers through multiobjective optimization to find a set of optimal tradeoffs between accuracy and any related cost. This approach allows find a suitable balance between all objectives regardless of the scenario. Experimental evaluations were done for Sound Environment Classification in modern battery-powered hearing aids by jointly optimising classification accuracy and computational costs.

A Multinomial Logistic Regression Approach for Arrhythmia Detection

Cardiovascular diseases are the leading causes on mortality in the world. Consequently, tools and methods providing useful and applicable insights into their assessment play a crucial role in the prediction and managements of specific heart conditions. In this article, we introduce a method based on multi-class Logistic Regression as a classifier to provide a powerful and accurate insight into cardiac arrhythmia, which is one of the predictors of serious vascular diseases. As suggested by our evaluation, this provides a robust, scalable, and accurate system, which can successfully tackle the challenges posed by the utilisation of big data in the medical sector.

Automated Fuzzy System Based on Feature Extraction and Selection for Opinion Classification Across Different Domains

Opinions have a relevant inuence on people’s behavior. The Internet and the Web have made it possible for people to share their opinions and for other people and organizations find out more about opinions and experiences from individuals. Still, opinions involve sentiments that are vague and imprecise textual descriptions. Hence, due to the nature of the data, Fuzzy Logic can be a promising approach. This paper proposes a method to automatically build a fuzzy system, based on features extracted and selected from documents, to perform classification of sentiment in opinions across different domains. Almost 60 features were extracted from documents and multiple feature selection algorithms were applied. Over the selected features, the Wang-Mendel (WM) method was used to generate fuzzy rules and classify documents. Variations on fuzzy set modeling, on the use of weights in the rules and on fuzzy inference mechanisms were considered. The classifier fuzzy system based achieved 71,25% of accuracy in a 10-fold cross-validation, comparable to a SVM classifier.

Modeling urban air pollution with optimized hierarchical fuzzy inference system

Environmental exposure assessments (EEA) and epidemiological studies require urban air pollution models with appropriate spatial and temporal resolutions. Uncertain available data and inflexible models can limit air pollution modeling techniques, particularly in under developing countries. This paper develops a hierarchical fuzzy inference system (HFIS) to model air pollution under different land use, transportation, and meteorological conditions. To improve performance, the system treats the issue as a large-scale and high-dimensional problem and develops the proposed model using a three-step approach. In the first step, a geospatial information system (GIS) and probabilistic methods are used to preprocess the data. In the second step, a hierarchical structure is generated based on the problem. In the third step, the accuracy and complexity of the model are simultaneously optimized with a multiple objective particle swarm optimization (MOPSO) algorithm. We examine the capabilities of the proposed model for predicting daily and annual mean PM2.5 and NO2 and compare the accuracy of the results with representative models from existing literature. The benefits provided by the model features, including probabilistic preprocessing, multi-objective optimization, and hierarchical structure, are precisely evaluated by comparing five different consecutive models in terms of accuracy and complexity criteria. Fivefold cross validation is used to assess the performance of the generated models. The respective average RMSEs and coefficients of determination (R (2)) for the test datasets using proposed model are as follows: daily PM2.5 = (8.13, 0.78), annual mean PM2.5 = (4.96, 0.80), daily NO2 = (5.63, 0.79), and annual mean NO2 = (2.89, 0.83). The obtained results demonstrate that the developed hierarchical fuzzy inference system can be utilized for modeling air pollution in EEA and epidemiological studies.

An Interpretability-Accuracy Tradeoff in Learning Parameters of Intuitionistic Fuzzy Rule-Based Systems

Parameter learning of Intuitionistic Fuzzy Rule-Based Systems (IFRBSs) is discussed and applied to medical diagnosis with intent of establishing a sound tradeoff between interpretability and accuracy. This study aims to improve the accuracy of IFRBSs without sacrificing its interpretability. This paper proposes an Objective Programming Method with an Interpretability-Accuracy tradeoff (OPMIA) to learn the parameters of IFRBSs by tuning the types of membership and non-membership functions and by adjusting adaptive factors and rule weights. The proposed method has been validated in the context of a medical diagnosis problem and a well-known publicly available auto-mpg data set. Furthermore, the proposed method is compared to Objective Programming Method not considering the interpretability (OPMNI) and Objective Programming Method based on Similarity Measure (OPMSM). The OPMIA helps achieve a sound a tradeoff between accuracy and interpretability and demonstrates its advantages over the other two methods.

A Genetic Algorithm Approach for Discovering Tuned Fuzzy Classification Rules with Intra- and Inter-Class Exceptions

Fuzzy rule-based systems (FRBSs) are proficient in dealing with cognitive uncertainties like vagueness and ambiguity imperative to real-world decision-making situations. Fuzzy classification rules (FCRs) based on fuzzy logic provide a framework for a flexible human-like reasoning involving linguistic variables. Appropriate membership functions (MFs) and suitable number of linguistic terms – according to actual distribution of data – are useful to strengthen the knowledge base (rule base [RB]+ data base [DB]) of FRBSs. An RB is expected to be accurate and interpretable, and a DB must contain appropriate fuzzy constructs (type of MFs, number of linguistic terms, and positioning of parameters of MFs) for the success of any FRBS. Moreover, it would be fascinating to know how a system behaves in some rare/exceptional circumstances and what action ought to be taken in situations where generalized rules cease to work. In this article, we propose a three-phased approach for discovery of FCRs augmented with intra- and inter-class exceptions. A pre-processing algorithm is suggested to tune DB in terms of the MFs and number of linguistic terms for each attribute of a data set in the first phase. The second phase discovers FCRs employing a genetic algorithm approach. Subsequently, intra- and inter-class exceptions are incorporated in the rules in the third phase. The proposed approach is illustrated on an example data set and further validated on six UCI machine learning repository data sets. The results show that the approach has been able to discover more accurate, interpretable, and interesting rules. The rules with intra-class exceptions tell us about the unique objects of a category, and rules with inter-class exceptions enable us to take a right decision in the exceptional circumstances.