Rule Reduction Research Articles

Simplification of ANFIS based on importance-confidence-similarity measures

Adaptive-network-based fuzzy inference system (ANFIS) is a well-known neuro-fuzzy system, which is widely implemented in nonlinear system approximation, prediction, control, and pattern classification. However, either too many fuzzy sets or fuzzy rules are indispensable to establish an accurate ANFIS for a high-dimensional system, which increases the complexity and limits the widespread application of the ANFIS. In this paper, an approach to simplify the ANFIS is developed based on importance-confidence-similarity (ICS) measures. The simplification process consists of selection of fuzzy rules with large importance-confidence (IC) values and reduction of similar fuzzy rules. The importance is defined as the average normalized firing strength of the rule. The confidence which reflects the approximation performance of a T-S fuzzy rule is estimated by the weighted orthogonal distance from the output data to the rule hyperplane (the consequent of a linear T-S fuzzy rule). The IC measures provide a comprehensive basis for the selection of the appropriate fuzzy rules. Furthermore, to the best of our knowledge, an approach to measure similarities of the T-S fuzzy rules including the rule premises and consequents is first proposed in this paper. We also develop a direct and simple method to remove the similar fuzzy rules based on a hierarchical tree and the IC measures. The proposed simplification approach is applied to Mackey-Glass chaotic time series prediction problem and the strip temperature increment prediction problem of the indirect-fired furnace of the annealing heating line. The results show that our proposed method achieves a balance between simplicity and accuracy of the ANFIS for the high-dimensional system.

Financial Constraints and Funding Solutions for Small Startups in Emerging Markets

This paper aims to analyse the challenges that small startups experience in emerging markets when accessing finance, the feasibility of non-traditional sources of funding, and the relevant policies that can help enhance funding. Intensified poverty and a lack of physical infrastructure that factors traditional banking solutions derive their effectiveness from means small startups in these areas have low access to credit, high interest rates, and underdeveloped venture capital markets. These constraints limit their advancement, expansion, and capacity to operate and maintain their businesses and thus are characterised by high failure rates. Others are; crowdfunding, microfinancing, and digital lending which can be considered as crucial methods used in bridging the gap in funding. Crowdfunding equalises capital access since Startups can issue funds directly from the public while microfinance offers small and flexible credit to nonmainstream business people. Digital lending comes with the assistance of technology which enables fast and convenient access to the loans and, therefore, supplements the funding challenge that startups with no credit pasts face. However, each of these solutions has restrictions in terms of fees, the maximum loan amount, and problems with regulation. This paper emphasises the need for appropriate policies to support the improvement of access to finance among startups. Other measures proposed are the development of credit guarantee schemes, reduction of bureaucratic rules and regulations, encouragement of FDI through tax incentives, and the use of technology in the provision of financial services. Besides, enlarging the financial literacy of intending and existing businessmen and enhancing the credit information infrastructure can very much enhance financing. From the study, it becomes clear that it will need to employ diversified strategies that put into practice various non-conventional funding paradigms alongside positive policies to facilitate a vibrant and progressive entrepreneurial culture within emerging markets.

A Fault Diagnosis Method for Autonomous Underwater Vehicles Based on Online Extended Belief Rule Base With Dynamic Rule Reduction

A Fault Diagnosis Method for Autonomous Underwater Vehicles Based on Online Extended Belief Rule Base With Dynamic Rule Reduction

A double inference engine belief rule base for oil pipeline leakage

Accurate prediction of oil pipeline leakage is important for energy security and environmental protection. The belief rule base (BRB) is a rule-based modeling approach that can make use of information with uncertainty to describe causal relationships in predicting oil pipeline leaks. Due to the complexity and uncertainty of oil pipeline systems, traditional BRB models produce a large rule base, which reduces the modeling capability and interpretability of BRB. Therefore, this paper proposes a double inference engine BRB (BRB-DI) model. Compared with the traditional BRB models, the new model diminishes the number of the rules from 56 to 25, while maintaining the accuracy. In the BRB-DI model, firstly, rule reduction is used to reduce the complexity of the model by comprehensively analyzing the importance of rules. Secondly, to ensure the completeness of the model rule base, a double inference engine consisting of the evidence reasoning (ER) algorithm and ER rule is proposed, and a new reasoning computation process is designed. Finally, an optimization algorithm based on projection covariance matrix adaptation evolution strategy (P-CMA-ES) is proposed to prevent the diminishing interpretability of the optimized model. In order to verify the effectiveness of the proposed method, an oil pipeline leakage prediction problem is studied as a numerical example.

On the influencing factors of AH shares linkage in China under new regulation policy of reduction

AH stock premium and its price linkage have always been the key problems of concern for scholars and investors. Taking the stock price yield of AH stock as the explanatory variable, this paper adopts the panel quantile model to probe into the linkage mechanism of AH stock price from five aspects such as risk difference, liquidity difference, demand elasticity difference, information asymmetry and new regulation policy of reduction. Due to the herding effect, information asymmetry has a negative effect on the H-share yield in the income and loss intervals, and a positive effect in the stable interval. The negative effect of liquidity difference on the H-share yield in the loss interval is converted into the positive effect in the income interval, but it is the most sensitive in high risk areas. The influence of demand elasticity on H-share yield is characterized by the typical characteristics of Chinese stock market “chasing after go up and killing cheapen” or “buying the winners”. Risk difference is not obvious because the behavior of risk lover and risk averter is offset by each other. The introduction of a 2017-year reduction in the new rules has significantly reduced the share price yield of H-shares. Finally, some suggestions are put forward for investors to invest and avoid risk in AH stocks and some specific policy implications for financial regulators are also provided based on these results.

An explainable decision model based on extended belief-rule-based systems to predict admission to the intensive care unit during COVID-19 breakout

Given the shortage of intensive care units (ICUs) due to the coronavirus disease (COVID-19), a prediction model is essential in ensuring the availability of ICU beds. However, several challenges, such as the importance of distinguishing indicators, efficiency of ICU admission records, and the explainability and effectiveness of the prediction model, hinder the effective prediction of ICU admissions. To mitigate these challenges, an explainable decision model that uses the extended belief rule-based system is introduced to predict ICU admission. First, an indicator extraction model is proposed to measure the importance of the various indicators and obtain representative indicators. Second, a Charnes, Cooper, and Rhodes (CCR) model is constructed to measure the efficiencies of the belief rules to achieve the compact structure of an extended belief rule base. Third, a new extended belief rule-based model, optimized by parameter optimization and domain division-based rule reduction, is developed to predict ICU admission. These procedures enable the explainable decision model to adapt to big data situation, offer explanations and realize high efficiency. Finally, a case study of ICU admission during a COVID-19 outbreak is conducted to demonstrate the implementation and effectiveness of the proposed model in a comparative analysis.

Fuzzy entropy DEMATEL inference system for accurate and efficient cardiovascular disease diagnosis

The global population is at risk from both communicable and non-communicable deadly diseases, including cardiovascular disease. Early detection and prevention of cardiovascular disease require an accurate self-detection model. Therefore, this study introduces a novel fuzzy entropy DEMATEL inference system for accurate self-detection of cardiovascular disease. It combines fuzzy DEMATEL, entropy, and Mamdani fuzzy inference, utilizing innovative strategies like attribute reduction, entropy-based clustering, influential factor selection, and rule reduction. The system achieves high accuracy (98.69%) and sensitivity (98.62%), outperforming existing methods. Validation includes satisfactory factor analysis, performance measures and statistical analysis, demonstrating its effectiveness in addressing complexity and prioritizing factors.

Fine-grained access control policy in blockchain-enabled edge computing

In the process of data sharing and interaction between untrusted intelligence edge devices, there are risks of misuse and tampering with the data. Hence, how to ensure data security and prevent unauthorized data access is the key problem to be solved in the cloud edge collaborative architecture. However, due to the heterogeneity of edge devices and the diversity of data sources, traditional access control methods cannot meet the requirements of the large-scale, distributed, dynamic and flexible expansion of edge intelligence applications. Therefore, this paper studies the fine-grained data access control method on the basis of smart contracts, and implements the attribute-based access control (ABAC) model through smart contracts. To improve the query efficiency of attributes and policies in the process of access request decision, an attribute management scheme is designed on the basis of the Bloom filter to achieve fast retrieval of attributes and policies. The decentralized fine-grained access control method is tested under the performance indicators such as retrieval delay, deployment delay, rule reduction rate and decision response time.

Multiple kernel learning-based rule reduction method for fuzzy modeling

As a model for complex processes, the Takagi-Sugeno fuzzy model (T-S fuzzy model) usually requires numerous rules, often leading to dimension disaster. To mitigate this problem while preserving modeling accuracy, a multiple kernel learning-based rule reduction method (MKLRR) is proposed here. First, several fuzzy sets in the original T-S fuzzy model are merged into a larger single set. Dynamic behavior in these big fuzzy sets is strongly nonlinear, and that between sets is often completely different, rendering ineffective the local linear model structure of traditional T-S fuzzy modeling. To solve this difficulty, we develop a multiple kernel learning strategy to project each big fuzzy set from a low-dimensional space into a high-dimensional space using the corresponding projection function. In each high-dimensional space, data has a linear relation which is easily represented by a linear model. The final model is constructed through fuzzy inference by combining all linear models from these high-dimensional spaces, greatly reducing the rule number and preserving modeling accuracy due to the good approximating ability of multiple kernel learning. Two solving strategies are developed to subsequently train the model. Additional analysis and proofs state the effectiveness of the proposed method. Case study indicates that the constructed model performs better with a fewer number of rules than the other methods with a greater number of rules, which effectively solves the problem of dimension disaster.

Comparative Analysis of Clustering Methods for Fuzzy Classifiers Simplification

One problem of fuzzy systems for classification tasks is the exponential growth of rule generation, which translates into excessively long processing times. Different proposals in the literature address this problem through exhaustive rule reduction techniques, which achieve competitive results compared to conventional classifiers by reducing their computational complexity as well. This paper proposes a methodology that is comprised of two stages: 1) a clustering technique that helps identify the structure of an initial low-accurate classifier and 2) a differential evolution parameter identification stage that takes that low-accurate classifier and refines it to obtain a high-accurate classifier. In this way, a comparative analysis among several clustering methods is performed, allowing the users to select a reduced rule set and avoid the use of traditional rule search algorithms. The results show that the Gaussian Mixture Model is the most suitable clustering technique to identify the structure of fuzzy classifiers, since it provides the corresponding fuzzy model with highly competitive classification performance compared to other state-of-the-art methods.

A Fuzzy Knowledge Graph Pairs-Based Application for Classification in Decision Making: Case Study of Preeclampsia Signs

Problems of preeclampsia sign diagnosis are mostly based on symptom data with the characteristics of data collected periodically in uncertain, ambiguous, and obstetrician opinions. To reduce the effects of preeclampsia, many studies have investigated the disease, prevention, and complication. Conventional fuzzy inference techniques can solve several diagnosis problems in health such as fuzzy inference systems (FIS), and Mamdani complex fuzzy inference systems with rule reduction (M-CFIS-R), however, the computation time is quite high. Recently, the research direction of approximate inference based on fuzzy knowledge graph (FKG) has been proposed in the M-CFIS-FKG model with the combination of regimens in traditional medicine and subclinical data gathered from medical records. The paper has presented a proposed model of FKG-Pairs3 to support patients’ disease diagnosis, together with doctors’ preferences in decision-making. The proposed model has been implemented in real-world applications for disease diagnosis in traditional medicine based on input data sets with vague information, quantified by doctor’s preferences. To validate the proposed model, it has been tested in a real-world case study of preeclampsia signs in a hospital for disease diagnosis with the traditional medicine approach. Experimental results show that the proposed model has demonstrated the model’s effectiveness in the decision-making of preeclampsia signs.

Rule reductions of decision formal context based on mixed information

Formal concept analysis is an effective tool in knowledge discovery. Nonetheless, it ignores the negative information, especially in rule extraction. And some applications require to mix positive and negative information for management and representation explicitly. This paper discusses an emerging method for rule reduction in formal contexts with mixed information by presenting a mixed decision rule based on a mixed concept lattice. Based on the mixed concept model, we discuss the relationships between mixed concept, formal concept and three-way concept, respectively. In this paper, to fully consider the mixed information, we construct a mixed decision rule. Moreover, to reduce the redundance of rule, a novel approach for weak-basis from the viewpoint of granular computing is designed to selection the core mixed decision rules and necessary mixed decision rules. Finally, the comparison of mixed decision rules and three-way decision rules is perfectly discussed, and case study is presented for the difference.

Variation of parental feeding practices during the COVID-2019 pandemic: a systematic review

Coronavirus disease 2019 (COVID-19) is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which is highly contagious and has resulted in a protracted pandemic. Infections caused by new coronavirus strains, primarily Delta and Omicron and currently highly prevalent globally. In response to the epidemic, countries, and cities implemented isolation and quarantine guidance, such as limiting social contact, which have affected the lifestyles and quality of life of the population. Parental feeding behaviors may vary as a result of factors such as prolonged home isolation of parents and children, lack of supplies during isolation, and stress. This study was designed to assess the available evidence and its implications for parental feeding practices in the context of COVID-19. We screened and reviewed research published in five electronic databases between 2020 and 2022, and eight studies met the selection criteria. Parents were observed to use a variety of feeding practices, including high levels of coercive control and reduction of rules and limits according to Vaughn’s food parenting constructs. The findings suggest that parental feeding practices are changing as a result of the pandemic and that more research is needed to further explore how to provide supportive feeding guidance to parents during emergencies in order to jointly promote child health.

RuleCOSI+: Rule extraction for interpreting classification tree ensembles

Despite the advent of novel neural network architectures, tree-based ensemble algorithms such as random forests and gradient boosting machines still prevail in many practical machine learning problems in manufacturing, financial, and medical domains. However, tree ensembles have the limitation that the internal decision mechanisms of complex models are difficult to understand. Therefore, we present a post-hoc interpretation approach for classification tree ensembles. The proposed method, RuleCOSI+, extracts simple rules from tree ensembles by greedily combining and simplifying their base trees. Compared with its previous version, RuleCOSI, this new version can be applied to both bagging (e.g., random forest, RF) and boosting ensembles (e.g., gradient boosting machines, GBM) and run much faster for ensembles with hundreds of trees. To assess the performance and applicability of the method, empirical experiments were conducted using two bagging algorithms and four gradient boosting algorithms over 33 datasets. RuleCOSI+ could generate the best classification rulesets in terms of F-measure together with RuleFit for RF and GBM models of the datasets among five ensemble simplification algorithms, but the rulesets of RuleCOSI+ had, on average, less than half the size of those of RuleFit. Moreover, RuleCOSI+ had the best antecedent uniqueness rate (“uniq”) among the five algorithms, and had also ranked high in the number of rules (“Nrules”) and the rule reduction rate (“redu”). In addition, the proposed method could reduce generalization errors in the simplified rulesets to obtain, on average, slightly better classification errors than original models of two bagging and three gradient boosting algorithms except CATBoost.

Rule reduction for control of a building cooling system using explainable AI

Although it is widely acknowledged that reinforcement learning (RL) can be beneficial for building control, many RL-based control actions remain unexplainable in the daily practice of facility managers. This paper reports a rule reduction framework using explainable RL to enhance the practicality of the control strategy. First, deep Q-learning was applied to explore the optimal control strategies of a parallel cooling system (ice-based thermal system + geothermal heat pump system) of an existing office building. A set of modularized and interconnected data-driven models was developed using ANNs for pretraining an artificial agent. After exploring the control strategies, the decision-making rules of the agent were reduced using a decision tree. The performance of the reduced-order rule-based control proved comparable to the complex and uninterpretable control strategy of deep Q-learning. The difference in energy savings between the two is marginal at 1.2%.

Rough set-based entropy measure with weighted density outlier detection method

Abstract The rough set theory is a powerful numerical model used to handle the impreciseness and ambiguity of data. Many existing multigranulation rough set models were derived from the multigranulation decision-theoretic rough set framework. The multigranulation rough set theory is very desirable in many practical applications such as high-dimensional knowledge discovery, distributional information systems, and multisource data processing. So far research works were carried out only for multigranulation rough sets in extraction, selection of features, reduction of data, decision rules, and pattern extraction. The proposed approach mainly focuses on anomaly detection in qualitative data with multiple granules. The approximations of the dataset will be derived through multiequivalence relation, and then, the rough set-based entropy measure with weighted density method is applied on every object and attribute. For detecting outliers, threshold value fixation is performed based on the estimated weight. The performance of the algorithm is evaluated and compared with existing outlier detection algorithms. Datasets such as breast cancer, chess, and car evaluation have been taken from the UCI repository to prove its efficiency and performance.

Slack and prices during Covid-19: Accounting for labor market participation

Strong labor force participation cyclicality during the Covid-19 pandemic has put further into question the capacity of standard Phillips Curve (PC) models to fully capture labor market cyclical conditions. In this paper, we jointly estimate natural unemployment and participation rates (i.e. compatible with constant inflation) through an augmented PC informed by structural labor market flows across employment, unemployment and inactivity. Focusing on Italy we find that, during the pandemic: (i) natural unemployment has remained unchanged, while natural participation has declined slightly, mostly due to a rise in retirement flows driven by a temporary reduction in pension eligibility rules; (ii) virtually all slack was accounted for by the participation margin, which added significant downward pressures to inflation dynamics.

Accounting and Financial Management Cost Accounting Integrating Rough Set Knowledge Recognition Algorithm

The main difference between rough set theory and some methods of uncertainty theory, such as probabilistic data mining, fuzzy set theory, and evidence‐based data mining, is that they do not require any prior knowledge beyond the data set being processed. This is also its advantage. At present, rough set theory has been well applied in artificial intelligence, knowledge discovery, pattern recognition, fault detection, and so on. According to the discovery model of classification knowledge, the attribute reduction of decision table, classification rule reduction, and classification algorithm under the condition of missing attribute are discussed. It tests the effectiveness of the knowledge recognition algorithm. The effectiveness of the algorithm proposed in this article reaches 87.6%, 84.4%, 94.97%, and 96.34%.

An Enhanced Exploration and Exploitation of Modified Grey Wolf Optimizer for Fuzzy Rules Reduction in Cloud Intrusion Detection System (CIDS)

Optimization problems in the Cloud Intrusion Detection System (CIDS) contain numerous conflicting objectives, uni-modal and multi-modal functions and the difficulty level varying between linear to non-linear limits. The Grey Wolf Optimizer (GWO) is a meta-heuristic technique that is implemented based on social behavior of wolves and their hunting behavior. Significant improvement in exploration and exploitation of the search space in GWO can be obtained by modifying the control parameter a. Works have already been carried out by many researchers by modifying the control parameter a in different manner to achieve better results. In a similar content, the authors of this study also modified the control parameter a and added a weight factor to the position of each wolf to attain the best possible results. Due to high demand for the cloud computing environment, intrusion detection in a cloud network plays a big role in maintaining the faith of the clients. Hence, CIDS is required to inspect the network packets to identify the abnormal behavior. For developing a system for cloud based IDS, the researchers created fuzzy rules to represent the relationship between the attributes and the nature of activity (normal or abnormal). The Modified Grey Wolf Optimizer (MGWO) algorithm is applied on eleven benchmark test functions and obtained good performance metrics. The results presented in this paper are promising; MGWO is used to reduce the fuzzy rules in developing fuzzy based CIDS. The performance of the proposed algorithm is compared with classical GWO, Particle Swarm Optimization (PSO), Cuckoo Search (CS) and variant of MGWO. The experimental results reveal that there is significant improvement in its performance.

An evolutionary algorithm based feature selection and fuzzy rule reduction technique for the prediction of skin cancer

SummaryIn current years, the death rate from skin cancers (SCs) tends to develop pretty. Various research verified that SC rank third as a deadliest disease, after breast and lung cancer. It will become vital to diagnose this malignancy at an early stage. The objective of this research is to mix machine learning and soft computing techniques to gain higher accuracy within the prediction of SC. To play out the exploration work, we utilized two data sets, one from “Save Life Hospital,” India, and the other is the UCI repository skin cancer data set. In this article, three meta‐heuristic algorithms, the FS_GA, the FS_PSO, and the FS_ACO, were used to select the best features from the data set provided to it. The AFRG_algorithm generates a set of fuzzy rules automatically and the RR_algorithm reduces certain fuzzy rules from the fuzzy system. For the SCC_dataset, the end accuracy obtained was 97.67%, 98.45%, and 99.22%. For the UCI_dataset, the end accuracy obtained was 98.81%, 99.72%, and 99.67%. Experimental results on the used datasets show that the proposed method strikingly improves the forecast exactitude of skin malignancy.