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

Residual Sketch Learning for a Feature-Importance-Based and Linguistically Interpretable Ensemble Classifier

Motivated by both the commonly used "from wholly coarse to locally fine" cognitive behavior and the recent finding that simple yet interpretable linear regression model should be a basic component of a classifier, a novel hybrid ensemble classifier called hybrid Takagi-Sugeno-Kang fuzzy classifier (H-TSK-FC) and its residual sketch learning (RSL) method are proposed. H-TSK-FC essentially shares the virtues of both deep and wide interpretable fuzzy classifiers and simultaneously has both feature-importance-based and linguistic-based interpretabilities. RSL method is featured as follows: 1) a global linear regression subclassifier on all original features of all training samples is generated quickly by the sparse representation-based linear regression subclassifier training procedure to identify/understand the importance of each feature and partition the output residuals of the incorrectly classified training samples into several residual sketches; 2) by using both the enhanced soft subspace clustering method (ESSC) for the linguistically interpretable antecedents of fuzzy rules and the least learning machine (LLM) for the consequents of fuzzy rules on residual sketches, several interpretable Takagi-Sugeno-Kang (TSK) fuzzy subclassifiers are stacked in parallel through residual sketches and accordingly generated to achieve local refinements; and 3) the final predictions are made to further enhance H-TSK-FC's generalization capability and decide which interpretable prediction route should be used by taking the minimal-distance-based priority for all the constructed subclassifiers. In contrast to existing deep or wide interpretable TSK fuzzy classifiers, benefiting from the use of feature-importance-based interpretability, H-TSK-FC has been experimentally witnessed to have faster running speed and better linguistic interpretability (i.e., fewer rules and/or TSK fuzzy subclassifiers and smaller model complexities) yet keep at least comparable generalization capability.

Constrained neuro fuzzy inference methodology for explainable personalised modelling with applications on gene expression data

Interpretable machine learning models for gene expression datasets are important for understanding the decision-making process of a classifier and gaining insights on the underlying molecular processes of genetic conditions. Interpretable models can potentially support early diagnosis before full disease manifestation. This is particularly important yet, challenging for mental health. We hypothesise this is due to extreme heterogeneity issues which may be overcome and explained by personalised modelling techniques. Thus far, most machine learning methods applied to gene expression datasets, including deep neural networks, lack personalised interpretability. This paper proposes a new methodology named personalised constrained neuro fuzzy inference (PCNFI) for learning personalised rules from high dimensional datasets which are structurally and semantically interpretable. Case studies on two mental health related datasets (schizophrenia and bipolar disorders) have shown that the relatively short and simple personalised fuzzy rules provided enhanced interpretability as well as better classification performance compared to other commonly used machine learning methods. Performance test on a cancer dataset also showed that PCNFI matches previous benchmarks. Insights from our approach also indicated the importance of two genes (ATRX and TSPAN2) as possible biomarkers for early differentiation of ultra-high risk, bipolar and healthy individuals. These genes are linked to cognitive ability and impulsive behaviour. Our findings suggest a significant starting point for further research into the biological role of cognitive and impulsivity-related differences. With potential applications across bio-medical research, the proposed PCNFI method is promising for diagnosis, prognosis, and the design of personalised treatment plans for better outcomes in the future.

Autonomous learning for fuzzy systems: a review

As one of the three pillars in computational intelligence, fuzzy systems are a powerful mathematical tool widely used for modelling nonlinear problems with uncertainties. Fuzzy systems take the form of linguistic IF-THEN fuzzy rules that are easy to understand for human. In this sense, fuzzy inference mechanisms have been developed to mimic human reasoning and decision-making. From a data analytic perspective, fuzzy systems provide an effective solution to build precise predictive models from imprecise data with great transparency and interpretability, thus facilitating a wide range of real-world applications. This paper presents a systematic review of modern methods for autonomously learning fuzzy systems from data, with an emphasis on the structure and parameter learning schemes of mainstream evolving, evolutionary, reinforcement learning-based fuzzy systems. The main purpose of this paper is to introduce the underlying concepts, underpinning methodologies, as well as outstanding performances of the state-of-the-art methods. It serves as a one-stop guide for readers learning the representative methodologies and foundations of fuzzy systems or who desire to apply fuzzy-based autonomous learning in other scientific disciplines and applied fields.

A new unmanned aerial vehicle intrusion detection method based on belief rule base with evidential reasoning

With the growing security demands in the public, civil and military fields, unmanned aerial vehicle (UAV) intrusion detection has attracted increasing attention. In view of the shortcomings of the current UAV intrusion detection model using Wi-Fi data traffic in terms of detection accuracy, sample size reduction, and model interpretability, this paper proposes a new detection algorithm for UAV intrusion. This paper presents an interpretable intrusion detection model for UAVs based on the belief rule base (BRB). BRB can effectively use various types of information to establish any nonlinear relationship between the model input and output. It can model and simulate any nonlinear model and optimize the model parameters. However, the rule combination explosion problem is encountered in BRB if there are too many attributes. Therefore, an evidential reasoning (ER) algorithm is proposed for solving this problem. By combining the capabilities of the ER and the BRB methodologies, a new evaluation model, named the EBRB-based model, is proposed here for predicting UAV intrusion detection, even in the case of a massive number of attributes. The global optimization of the model is ensured. A new interpretable and globally optimized UAV intrusion detection model is proposed, which is the main contribution of this paper. An experimental case is used to demonstrate the implementation and application of the proposed UAV intrusion detection method.

Evolving fuzzy neural classifier that integrates uncertainty from human-expert feedback

Evolving fuzzy neural networks are models capable of solving complex problems in a wide variety of contexts. In general, the quality of the data evaluated by a model has a direct impact on the quality of the results. Some procedures can generate uncertainty during data collection, which can be identified by experts to choose more suitable forms of model training. This paper proposes the integration of expert input on labeling uncertainty into evolving fuzzy neural classifiers (EFNC) in an approach called EFNC-U. Uncertainty is considered in class label input provided by experts, who may not be entirely confident in their labeling or who may have limited experience with the application scenario for which the data is processed. Further, we aimed to create highly interpretable fuzzy classification rules to gain a better understanding of the process and thus to enable the user to elicit new knowledge from the model. To prove our technique, we performed binary pattern classification tests within two application scenarios, cyber invasion and fraud detection in auctions. By explicitly considering class label uncertainty in the update process of the EFNC-U, improved accuracy trend lines were achieved compared to fully (and blindly) updating the classifiers with uncertain data. Integration of (simulated) labeling uncertainty smaller than 20% led to similar accuracy trends as using the original streams (unaffected by uncertainty). This demonstrates the robustness of our approach up to this uncertainty level. Finally, interpretable rules were elicited for a particular application (auction fraud identification) with reduced (and thus readable) antecedent lengths and with certainty values in the consequent class labels. Additionally, an average expected uncertainty of the rules were elicited based on the uncertainty levels in those samples which formed the corresponding rules.

An Explainable Evolving Fuzzy Neural Network to Predict the k Barriers for Intrusion Detection Using a Wireless Sensor Network

Evolving fuzzy neural networks have the adaptive capacity to solve complex problems by interpreting them. This is due to the fact that this type of approach provides valuable insights that facilitate understanding the behavior of the problem being analyzed, because they can extract knowledge from a set of investigated data. Thus, this work proposes applying an evolving fuzzy neural network capable of solving data stream regression problems with considerable interpretability. The dataset is based on a necessary prediction of k barriers with wireless sensors to identify unauthorized persons entering a protected territory. Our method was empirically compared with state-of-the-art evolving methods, showing significantly lower RMSE values for separate test data sets and also lower accumulated mean absolute errors (MAEs) when evaluating the methods in a stream-based interleaved-predict-and-then-update procedure. In addition, the model could offer relevant information in terms of interpretable fuzzy rules, allowing an explainable evaluation of the regression problems contained in the data streams.

Review on Interpretable Machine Learning in Smart Grid

In recent years, machine learning, especially deep learning, has developed rapidly and has shown remarkable performance in many tasks of the smart grid field. The representation ability of machine learning algorithms is greatly improved, but with the increase of model complexity, the interpretability of machine learning algorithms is worse. The smart grid is a critical infrastructure area, so machine learning models involving it must be interpretable in order to increase user trust and improve system reliability. Unfortunately, the black-box nature of most machine learning models remains unresolved, and many decisions of intelligent systems still lack explanation. In this paper, we elaborate on the definition, motivations, properties, and classification of interpretability. In addition, we review the relevant literature addressing interpretability for smart grid applications. Finally, we discuss the future research directions of interpretable machine learning in the smart grid.

A Fuzzy Strategy to Eliminate Uncertainty in Grading Positive Tuberculosis

Sputum smear microscopic examination is an effective, fast, and low-cost technique that is highly specific in areas with a high prevalence of pulmonary tuberculosis. Since manual screening needs trained pathologist in high prevalence zones, the possibility of deploying adequate technicians during the epidemic sessions would be impractical. This condition can cause overburdening and fatigue of working technicians which may tend to reduce the potential efficiency of Tuberculosis (TB) diagnosis. Hence, automation of sputum inspection is the most appropriate aspect in TB outbreak zones to maximize the detection ability. Sputum collection, smear preparing, staining, interpreting smears, and reporting of TB severity are all part of the diagnosis of tuberculosis. This study has analyzed the risk of automating TB severity grading. According to the findings of the analysis, numerous TB-positive cases do not fit into the standard TB severity grade while applying direct rule-driven strategy. The manual investigation, on the other hand, arbitrarily labels the TB grade on those cases. To counter the risk of automation, a fuzzy-based Tuberculosis Severity Level Categorizing Algorithm (TSLCA) is introduced to eliminate uncertainty in classifying the level of TB infection. TSLCA introduces the weight factors, which are dependent on the existence of maximum number of Acid-Fast Bacilli (AFB) per microscopic Field of View (FOV). The fuzzification and defuzzification operations are carried out using the triangular membership function. In addition, the [Formula: see text]-cut approach is used to eliminate the ambiguity in TB severity grading. Several uncertain TB microscopy screening reports are tested using the proposed TSLCA. Based on the experimental results, it is observed that the TB grading by TSLCA is consistent, error-free, significant and fits exactly into the standard criterion. As a result of the proposed TSLCA, the uncertainty of grading is eliminated and the reliability of tuberculosis diagnosis is ensured when adapting automatic diagnosis.

Business Intelligence in Airline Passenger Satisfaction Study—A Fuzzy-Genetic Approach with Optimized Interpretability-Accuracy Trade-Off

The main objective and contribution of this paper is the application of our knowledge-discovery business-intelligence technique (fuzzy rule-based classification systems) characterized by genetically optimized interpretability-accuracy trade-off (using multi-objective evolutionary optimization algorithms) to decision support related to airline passenger satisfaction problems. Recently published and accessible at Kaggle’s repository airline passengers satisfaction data set containing 259,760 records is used in our experiments. A comparison of our approach with an alternative method (using SAS-system’s accuracy-oriented prediction tools to determine the attribute importance hierarchy) is also performed showing the advantages of our method in terms of: (i) discovering the actual hierarchy of attribute significance for passenger satisfaction and (ii) knowledge-discovery system’s interpretability-accuracy trade-off optimization. The main results and findings of our work include: (i) an introduction of the modern fuzzy-genetic business-intelligence solution characterized both by high interpretability and high accuracy to the airline passenger satisfaction decision support, (ii) an analysis of the effect of possible "overlapping" of some input attributes over the other ones in order to discover the real hierarchy of influence of particular input attributes upon the airline passengers satisfaction, and (iii) an extended cross-validation experiment confirming high effectiveness of our approach for different learning-test splits of the data set considered.

Differential diagnosis of asthma and restrictive respiratory diseases by combining forced oscillation measurements, machine learning and neuro-fuzzy classifiers

To design machine learning classifiers to facilitate the clinical use and increase the accuracy of the forced oscillation technique (FOT) in the differential diagnosis of patients with asthma and restrictive respiratory diseases. FOT and spirometric exams were performed in 97 individuals, including controls (n = 20), asthmatic patients (n = 38), and restrictive (n = 39) patients. The first experiment of this study showed that the best FOT parameter was the resonance frequency, providing moderate accuracy (AUC = 0.87). In the second experiment, a neuro-fuzzy classifier and different supervised machine learning techniques were investigated, including k-nearest neighbors, random forests, AdaBoost with decision trees, and support vector machines with a radial basis kernel. All classifiers achieved high accuracy (AUC ≥ 0.9) in the differentiation between patient groups. In the third and fourth experiments, the use of different feature selection techniques allowed us to achieve high accuracy with only three FOT parameters. In addition, the neuro-fuzzy classifier also provided rules to explain the classification. Neuro-fuzzy and machine learning classifiers can aid in the differential diagnosis of patients with asthma and restrictive respiratory diseases. They can assist clinicians as a support system providing accurate diagnostic options.

A Modern Data-Mining Approach Based on Genetically Optimized Fuzzy Systems for Interpretable and Accurate Smart-Grid Stability Prediction

The main objective and contribution of this paper was/is the application of our knowledge-based data-mining approach (a fuzzy rule-based classification system) characterized by a genetically optimized interpretability-accuracy trade-off (by means of multi-objective evolutionary optimization algorithms) for transparent and accurate prediction of decentral smart grid control (DSGC) stability. In particular, we aim at uncovering the hierarchy of influence of particular input attributes upon the DSGC stability. Moreover, we also analyze the effect of possible "overlapping" of some input attributes over the other ones from the DSGC-stability perspective. The recently published and available at the UCI Database Repository Electrical Grid Stability Simulated Data Set and its input-aggregate-based concise version were used in our experiments. A comparison with 39 alternative approaches was also performed, demonstrating the advantages of our approach in terms of: (i) interpretable and accurate fuzzy rule-based DSGC-stability prediction and (ii) uncovering the hierarchy of DSGC-system’s attribute significance.

SK-MOEFS: A Library in Python for Designing Accurate and Explainable Fuzzy Models

Recently, the explainability of Artificial Intelligence (AI) models and algorithms is becoming an important requirement in real-world applications. Indeed, although AI allows us to address and solve very difficult and complicated problems, AI-based tools act as a black box and, usually, do not explain how/why/when a specific decision has been taken. Among AI models, Fuzzy Rule-Based Systems (FRBSs) are recognized world-wide as transparent and interpretable tools: they can provide explanations in terms of linguistic rules. Moreover, FRBSs may achieve accuracy comparable to those achieved by less transparent models, such as neural networks and statistical models. In this work, we introduce SK-MOEFS (acronym of SciKit-Multi Objective Evolutionary Fuzzy System), a new Python library that allows the user to easily and quickly design FRBSs, employing Multi-Objective Evolutionary Algorithms. Indeed, a set of FRBSs, characterized by different trade-offs between their accuracy and their explainability, can be generated by SK-MOEFS. The user, then, will be able to select the most suitable model for his/her specific application.

Performance and Interpretability in Fuzzy Logic Systems – Can We Have Both?

Fuzzy Logic Systems can provide a good level of interpretability and may provide a key building block as part of a growing interest in explainable AI. In practice, the level of interpretability of a given fuzzy logic system is dependent on how well its key components, namely, its rule base and its antecedent and consequent fuzzy sets are understood. The latter poses an interesting problem from an optimisation point of view – if we apply optimisation techniques to optimise the parameters of the fuzzy logic system, we may achieve better performance (e.g. prediction), however at the cost of poorer interpretability. In this paper, we build on recent work in non-singleton fuzzification which is designed to model noise and uncertainty ‘where it arises’, limiting any optimisation impact to the fuzzification stage. We explore the potential of such systems to deliver good performance in varying-noise environments by contrasting one example framework - ADONiS, with ANFIS, a traditional optimisation approach designed to tune all fuzzy sets. Within the context of time series prediction, we contrast the behaviour and performance of both approaches with a view to inform future research aimed at developing fuzzy logic systems designed to deliver both – high performance and high interpretability.

User Centered Neuro-Fuzzy Energy Management Through Semantic-Based Optimization

This paper presents a cloud-based building energy management system, underpinned by semantic middleware, that integrates an enhanced sensor network with advanced analytics, accessible through an intuitive Web-based user interface. The proposed solution is described in terms of its three key layers: 1) user interface; 2) intelligence; and 3) interoperability. The system's intelligence is derived from simulation-based optimized rules, historical sensor data mining, and a fuzzy reasoner. The solution enables interoperability through a semantic knowledge base, which also contributes intelligence through reasoning and inference abilities, and which are enhanced through intelligent rules. Finally, building energy performance monitoring is delivered alongside optimized rule suggestions and a negotiation process in a 3-D Web-based interface using WebGL. The solution has been validated in a real pilot building to illustrate the strength of the approach, where it has shown over 25% energy savings. The relevance of this paper in the field is discussed, and it is argued that the proposed solution is mature enough for testing across further buildings.

Local-Density-Based Optimal Granulation and Manifold Information Granule Description

Constructing information granules (IGs) has been of significant interest to the discipline of granular computing. The principle of justifiable granularity has been proposed to guide the design of IGs, opening an avenue of pursuits of building IGs carried out on a basis of well-defined and intuitively appealing principles. However, how to improve the efficiency and accuracy of the resulting constructs is an open issue. In this paper, we present a local-density-based optimal granulation model (LoDOG), exhibiting evident advantages: 1) it can detect arbitrarily-shaped IGs and 2) it finds the optimal granulation solutions with O(N) complexity, once the leading tree structure has been constructed. We describe IGs of arbitrary shapes using a small collection of landmark points positioned on the skeleton of the underlying manifold, which contribute to approximate reconstruction capabilities of the original dataset. A dissimilarity metric is developed to evaluate the quality of the obtained reconstruction. The interpretability of LoDOG IGs is discussed. Theoretical analysis and empirical evaluations are covered to demonstrate the effectiveness of LoDOG and the manifold description.