If-then Rules Research Articles

Towards explicit representation of an artificial neural network model: Comparison of two artificial neural network rule extraction approaches

In the quest for interpretable models, two versions of a neural network rule extraction algorithm were proposed and compared. The two algorithms are called the Piece-Wise Linear Artificial Neural Network (PWL-ANN) and enhanced Piece-Wise Linear Artificial Neural Network (enhanced PWL-ANN) algorithms. The PWL-ANN algorithm is a decomposition artificial neural network (ANN) rule extraction algorithm, and the enhanced PWL-ANN algorithm improves upon the PWL-ANN algorithm and extracts multiple linear regression equations from a trained ANN model by approximating the hidden sigmoid activation functions using N-piece linear equations. In doing so, the algorithm provides interpretable models from the originally trained opaque ANN models. A detailed application case study illustrates how the generated enhanced-PWL-ANN models can provide understandable IF-THEN rules about a problem domain. Comparison of the results generated by the two versions of the PWL-ANN algorithm showed that in comparison to the PWL-ANN models, the enhanced-PWL-ANN models support improved fidelities to the originally trained ANN models. The results also showed that more concise rule sets could be generated using the enhanced-PWL-ANN algorithm. If a more simplified set of rules is desired, the enhanced-PWL-ANN algorithm can be combined with the decision tree approach. Potential application of the algorithms to domains related to petroleum engineering can help enhance understanding of the problems.

Formalizing GDPR Provisions in Reified I/O Logic: The DAPRECO Knowledge Base

The DAPRECO knowledge base is the main outcome of the interdisciplinary project bearing the same name ( https://www.fnr.lu/projects/data-protection-regulation-compliance ). It is a repository of rules written in LegalRuleML, an XML formalism designed to be a standard for representing the semantic and logical content of legal documents. The rules represent the provisions of the General Data Protection Regulation (GDPR), the new Regulation that is significantly affecting the digital market in the European Union and beyond. The DAPRECO knowledge base builds upon the Privacy Ontology (PrOnto) (Palmirani et al in Proceedings of the 7th international conference on electronic government and the information systems perspective: technology-enabled innovation for democracy, government and governance, 2018c), which provides a model for the legal concepts involved in the GDPR, by adding a further layer of constraints in the form of if-then rules, referring either to standard first order logic implications or to deontic statements. If-then rules are formalized in reified Input/Output logic (Robaldo and Sun in J Log Comput 7, 2017) and then codified in LegalRuleML. Reified Input/Output logic is an application of standard Input/Output logic for legal reasoning, in which Input/Output logic is combined with the reification-based approach in Hobbs and Gordon (A formal theory of commonsense psychology, how people think people think. Cambridge University Press, Cambridge, 2017). The DAPRECO knowledge base is then a case study for reified Input/Output logic, and it shows that the formalism indeed appears to be a good candidate to effectively formalize, via uniform and simple (flat) representations, complex linguistic/deontic phenomena that may be found in legal texts. To date, the DAPRECO knowledge base is the biggest knowledge base in LegalRuleML and Input/Output logic freely available online ( https://github.com/dapreco/daprecokb/blob/master/gdpr/rioKB_GDPR.xml ).

A fuzzy logic based approach for prediction of basal cell carcinoma and squamous cell carcinoma among the data of skin cancer

INTRODUCTION: Both basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) is a type of skin malignancy which are deadly in nature. Although both can cause a serious setback for the human body, SCC is most dangerous as per human life is concerned.OBJECTIVES: It is necessary to spot out the cases of SCC and BCC among various data of skin cancer. In this research, the same is spotted out with the help of the fuzzy logic system.METHODS: At first, the membership function is constructed from the input data provided. Then based on the generated membership function, a set of fuzzy if-then rules are created. The outliers from the source data are also removed before the generation of the fuzzy membership function. Based on the fuzzy if-then rules, the decision, whether the case is of basal cell or squamous cell carcinoma is taken.RESULTS: The output prediction by the system is compared with the actual pathological report of the patients. The comparison provides an accuracy of 94.94 percent.CONCLUSION: This paper introduces a new technique to predict the existence of multiple cancers (either basal cell or squamous cell) from the mixed cancer dataset. The outcomes of the examination on the source dataset show that the proposed system achieved extraordinary desire accuracy for the portrayal of squamous cell carcinoma and basal cell carcinoma.

Pool Water Acidity Gauge Using Fuzzy Mamdani Method

Swimming pool is one of the means of sports and play that is fun. Swimming pools must also be considered clean. In order to be replaced with new water. The goal is so that pool users are not infected with infectious diseases due to the dirty water inside the pool. The Mamdani method is the most common method when discussing fuzzy methodology. It was Ebrahim Mamdani who first proposed this method in 1975 when building a steam engine and boiler control system. Mamdani uses a collection of IF-THEN rules obtained from experienced operators / experts. Mamdani’s work is actually based on the article “The Father of Fuzzy, Lotfi A. Zadeh: fuzzy algorithms for complex systems and decision processes.

Mamdani Fuzzy Inference System using Three Parameters for Flood Disaster Forecasting in Bandung region

This study aims to investigate the flood events forecast by applying the Mamdani fuzzy inference system and rule-based reasoning. For input and output linguistic variable of the model based on three parameters such as population density, altitude, as well the rainfall. Each variable has three to four membership functions. The data used come from the Statistic Bureau of Indonesia, National Disaster Management Authority, and Meteorology and Climatology Center of Indonesia for 2014 to present. A different model for each region was constructed depending on available data sets. We determine based our model on forty-eight fuzzy IF-THEN rules and fuzzy reasoning. The output variable which has three membership function takes from 220 to 402 as representative flooding level warning such as normal, alert, and danger. The results of the study have high agreements with recorded data in on site with increases the output values toward the real-time flood event. Therefore, if that area will be used for living, this will have risks that flood always is a threat to those areas.

Implementation of fuzzy tsukamoto method in decision support system of journal acceptance

Journal acceptance is a difficult problem to solve since in the practice it involves some reviewers who can produce different decisions from various perspectives. Therefore, a decision support system is needed to assist the reviewers to decide the acceptance of the paper. This study was purposed to develop the decision support system using the Fuzzy Tsukamoto method for journal acceptance. The Fuzzy Tsukamoto method described the relationship between input and output of the system by using a set of fuzzy if-then rules. From the comparison results, It is obtained that the accuracy from the result of comparison of the manual method, expert decision, and DSS of journal acceptance using Fuzzy Tsukamoto Method is 95% with 5% errors. Based on the results of the accuracy and error, it shows that the DSS of journal acceptance using the Fuzzy Tsukamoto Method is accurate and has high precision.

Type 2 Fuzzy Inference-Based Time Series Model

Fuzzy techniques have been suggested as useful method for forecasting performance. However, its dependency on experts’ knowledge causes difficulties in information extraction and data collection. Therefore, to overcome the difficulties, this research proposed a new type 2 fuzzy time series (T2FTS) forecasting model. The T2FTS model was used to exploit more information in time series forecasting. The concepts of sliding window method (SWM) and fuzzy rule-based systems (FRBS) were incorporated in the utilization of T2FTS to obtain forecasting values. A sliding window method was proposed to find a proper and systematic measurement for predicting the number of class intervals. Furthermore, the weighted subsethood-based algorithm was applied in developing fuzzy IF–THEN rules, where it was later used to perform forecasting. This approach provides inferences based on how people think and make judgments. In this research, the data sets from previous studies of crude palm oil prices were used to further analyze and validate the proposed model. With suitable class intervals and fuzzy rules generated, the forecasting values obtained were more precise and closer to the actual values. The findings of this paper proved that the proposed forecasting method could be used as an alternative for improved forecasting of sustainable crude palm oil prices.

Fuzzy Integrated Ontology Model of Dynamic Learner Profiling

In the context of lifelong learning, learner profile has emerged as a feasible model that support and promote the provision of lifelong learning opportunities. Learner profile describes the attributes and outcomes of education in a learning system. It includes information on learner’s gender, skills, education, interest, learning preferences, learning style, etc. This paper proposes an approach to construct a fuzzy based semantic learner profile in the promising technology of semantic web by using the concept of ontology and use it for the reasoning of learner preferences. The approach starts with the collection of learner’s static and dynamic data. The dynamic data of learner particularly learner interest and learning style are extracted by weblog analysis and using algorithms such as semantic based representation using WordNet and modified decision tree classifier with strong rules based on Felder-Silverman learning style model. The retrieved data is then used to construct learner profile using ontology in which automatic learner profile updating is obtained using ontology based semantic similarity algorithm. Finally to achieve semantic retrieval from learner profile ontology, fuzzy concepts such as fuzzy linguistic variable and fuzzy IF THEN rules are applied. Fuzzy linguistic variable facilitate semantic retrieval and more specific classification from learner profile ontology and fuzzy IF THEN rules predict the learning preference of new students based on the forward chaining reasoning process implemented in the existing ontology model. The final representation of semantic fuzzy ontology based learner profile improves the performance of tasks such as classification, semantic retrieval and prediction of learning preference to the new learners. The case study is conducted for the real-time learners involved in studying the courses registered in Moodle Learning Management System. The experiments were performed with NetBeans IDE, Jena framework and Protégé 4.2 beta editor. The experiments confirm that the proposed learner profile is a good representation of the learner's preferences.

A faster convergence and concise interpretability TSK fuzzy classifier deep-wide-based integrated learning

Hierarchical TSK fuzzy system was proposed to approach the exponential growth of IF-THEN rules which named “fuzzy rule explosion”. However, it could not get better performance in few layers for instability of TSK fuzzy system, such that hierarchical TSK fuzzy system suffers from bad interpretability and slow convergence along with too much layers. To get a better solution, this study employs a faster convergence and concise interpretability TSK fuzzy classifier deep-wide-based integrated learning (FCCI-TSK) which has a wide structure to adopt several ensemble units learning in a meantime, and the best performer will be picked up to transfer its learning knowledge to next layer with the help of stacked generalization principle. The ensemble units are integrated by negative correlation learning (NCL). FCCI-TSK adjusts the input of the next layer with a better guidance such that it can quicken the speed of convergence and reduce the number of layers. Besides, leading with guidance, it can achieve higher accuracy and better interpretability with more simple structure. The contributions of this study include: (1) To enhance the performance of fuzzy classifier, we mix NCL and stacked generalization principle together in FCCI-TSK; (2) To overcome the phenomenon of “fuzzy rule explosion”, we adopt deep-wide integrated learning and information discarding to accelerate convergence and obtain concise interpretability in the meantime. Comparing with other 11 algorithms, the results on twelve UCI datasets show that FCCI-TSK has the best performance overall and the convergence of FCCI-TSK is also examined.

Knowledge-based reinforcement learning controller with fuzzy-rule network: experimental validation

A model-free controller for a general class of output feedback nonlinear discrete-time systems is established by action-critic networks and reinforcement learning with human knowledge based on IF–THEN rules. The action network is designed by a single input fuzzy-rules emulated network with the set of IF–THEN rules utilized by the relation between control effort and plant’s output such as IF the output is high THEN the control effort should be reduced. The critic network is constructed by a multi-input FREN (MiFREN) for estimating an unknown long-term cost function. The set of IF–THEN rules for MiFREN is defined by the general knowledge of optimization such that IF the quadratic values of control effort and tracking error are high THEN the cost function should be high. The convergence of tracking error and bounded external signals can be guaranteed by Lyapunov direct method under general assumptions which are reasonable for practical plants. A computer simulation system is firstly provided to demonstrate the design method and the performance of the proposed controller. Furthermore, an experimental system with the prototype of DC-motor current control is conducted to show the effectiveness of the control scheme.

Data driven controller based on fuzzy rule adaptive network: with experimental validation

PurposeThe purpose of this paper is to design an online-data driven adaptive control scheme based on fuzzy rules emulated network (FREN) for a class of unknown nonlinear discrete-time systems.Design/methodology/approachBy using the input-output characteristic curve of controlled plant and the set of IF-THEN rules based on human knowledge inspiration, the adaptive controller is established by an adaptive FREN. The learning algorithm is established with convergence proof of the closed-loop system and controller’s parameters are directly designed by experimental data.FindingsThe convergence of tracking error is verified by the theoretical results and the experimental systems. The experimental systems and comparison results show that the proposed controller and its design procedure based on input-output data can achieve superior performance.Practical implicationsThe theoretical aspect and experimental systems with the light-emitting diode (LED) current control and the robotic system prove that the proposed controller can be designed by using only input-output data of the controlled plants when the tracking error can be affirmed the convergence.Originality/valueThe proposed controller has been theoretically developed and used through experimental systems by using only input-output data of the controlled plant. The novel design procedure has been proposed by using the input-output characteristic curve for both positive and negative control directions.

FES-Induced Cycling in Complete SCI: A Simpler Control Method Based on Inertial Sensors.

This article introduces a novel approach for a functional electrical stimulation (FES) controller intended for FES-induced cycling based on inertial measurement units (IMUs). This study aims at simplifying the design of electrical stimulation timing patterns while providing a method that can be adapted to different users and devices. In most of studies and commercial devices, the crank angle is used as an input to trigger stimulation onset. We propose instead to use thigh inclination as the reference information to build stimulation timing patterns. The tilting angles of both thighs are estimated from one inertial sensor located above each knee. An IF–THEN rule algorithm detects, online and automatically, the thigh peak angles in order to start and stop the stimulation of quadriceps muscles, depending on these events. One participant with complete paraplegia was included and was able to propel a recumbent trike using the proposed approach after a very short setting time. This new modality opens the way for a simpler and user-friendly method to automatically design FES-induced cycling stimulation patterns, adapted to clinical use, for multiple bike geometries and user morphologies.

Fuzzy Rules Emulated Discrete-Time Controller Based on Plant’s Input–Output Association

In this work, an adaptive controller is designed by an adaptive network called fuzzy rules emulated network (FREN) for a class of nonlinear discrete-time systems when the mathematical model is completely neglected. The characteristic curve between the plant’s input and output is drawn by results of the experimental setup. The design parameters are determined by the characteristic curve and utilized to design the controller. The network structure of FREN controller is established by the set of human knowledge-based IF-THEN rules according to the relation between the plant’s input and output. The theoretical result is provided to guarantee the convergence of tracking error of the learning algorithm. Furthermore, experimental systems with the LED current control and the robotic system are constructed to validate the performance of the proposed controller.

InstructableCrowd: Creating IF-THEN Rules for Smartphones via Conversations with the Crowd

Natural language interfaces have become a common part of modern digital life. Chatbots utilize text-based conversations to communicate with users; personal assistants on smartphones such as Google Assistant take direct speech commands from their users; and speech-controlled devices such as Amazon Echo use voice as their only input mode. In this paper, we introduce InstructableCrowd, a crowd-powered system that allows users to program their devices via conversation. The user verbally expresses a problem to the system, in which a group of crowd workers collectively respond and program relevant multi-part IF-THEN rules to help the user. The IF-THEN rules generated by InstructableCrowd connect relevant sensor combinations (e.g., location, weather, device acceleration, etc.) to useful effectors (e.g., text messages, device alarms, etc.). Our study showed that non-programmers can use the conversational interface of InstructableCrowd to create IF-THEN rules that have similar quality compared with the rules created manually. InstructableCrowd generally illustrates how users may converse with their devices, not only to trigger simple voice commands, but also to personalize their increasingly powerful and complicated devices.

InstructableCrowd: Creating IF-THEN Rules for Smartphones via Conversations with the Crowd

Natural language interfaces have become a common part of modern digital life. Chatbots utilize text-based conversations to communicate with users; personal assistants on smartphones such as Google Assistant take direct speech commands from their users; and speech-controlled devices such as Amazon Echo use voice as their only input mode. In this paper, we introduce InstructableCrowd, a crowd-powered system that allows users to program their devices via conversation. The user verbally expresses a problem to the system, in which a group of crowd workers collectively respond and program relevant multi-part IF-THEN rules to help the user. The IF-THEN rules generated by InstructableCrowd connect relevant sensor combinations (e.g., location, weather, device acceleration, etc.) to useful effectors (e.g., text messages, device alarms, etc.). Our study showed that non-programmers can use the conversational interface of InstructableCrowd to create IF-THEN rules that have similar quality compared with the rules created manually. InstructableCrowd generally illustrates how users may converse with their devices, not only to trigger simple voice commands, but also to personalize their increasingly powerful and complicated devices.

A chaotic iterative fuzzy modeling of circulating a simple sentence

In this paper, we propose a new model to describe variations in interpretation and perception of a simple sentence by different people. To show the understandability of a simple sentence in the prediction of future situations, the meaning of a sentence is modeled as a fuzzy if-then rule, and the fuzzy model is investigated in an iterative process. The main goal of the paper is modeling a linguistic rule. This is done using an if-then rule and its variation through one person to another. The model predicts that the interpretation reaching the final person in the following years can be chaotic and thus unpredictable.

RecRules

Nowadays, end users can personalize their smart devices and web applications by defining or reusing IF-THEN rules through dedicated End-User Development (EUD) tools. Despite apparent simplicity, such tools present their own set of issues. The emerging and increasing complexity of the Internet of Things, for example, is barely taken into account, and the number of possible combinations between triggers and actions of different smart devices and web applications is continuously growing. Such a large design space makes end-user personalization a complex task for non-programmers, and motivates the need of assisting users in easily discovering and managing rules and functionality, e.g., through recommendation techniques. In this article, we tackle the emerging problem of recommending IF-THEN rules to end users by presenting RecRules , a hybrid and semantic recommendation system. Through a mixed content and collaborative approach, the goal of RecRules is to recommend by functionality : it suggests rules based on their final purposes, thus overcoming details like manufacturers and brands. The algorithm uses a semantic reasoning process to enrich rules with semantic information, with the aim of uncovering hidden connections between rules in terms of shared functionality. Then, it builds a collaborative semantic graph, and it exploits different types of path-based features to train a learning to rank algorithm and compute top-N recommendations. We evaluate RecRules through different experiments on real user data extracted from IFTTT, one of the most popular EUD tools. Results are promising: they show the effectiveness of our approach with respect to other state-of-the-art algorithms and open the way for a new class of recommender systems for EUD that take into account the actual functionality needed by end users.

Achieving More Stringent Levels of Comfort via an Adaptive Fuzzy Controller Optimized by the Gravitational Search Algorithm for a Half-Body Car Model

An optimal adaptive fuzzy controller is designed to achieve more stringent levels of comfort for a half-body car model. This aim will be fulfilled by reducing road disturbances and decreasing the acceleration of the body. The proposed controller consists of two adaptive fuzzy controllers with two fuzzy systems. Each one has two inputs, one output and twenty five linguistic fuzzy IF-THEN rules. Every input has five Gaussian membership functions and uses the product inference engine, singleton fuzzifier and the centre average defuzzifier. In order to determine the optimal parameters for the Adaptive Fuzzy Controller (AFC), the Gravitational Search Algorithm (GSA) is applied. The relative displacement between spring mass and tire, along with the acceleration of the body, are the two objective functions being applied in the optimization algorithm. The results illustrate the superiority of the proposed optimal adaptive fuzzy controller in comparison with traditional controllers.

APPLICATION OF MAMDANI FUZZY METHOD TO DETERMINE ROUND BREAD PRODUCTION AT PT VANESSA BAKERY

Mamdani Fuzzy is a method that interprets input values and makes conclusions based on IF-THEN rules and producing the output. In this research Mamdani fuzzy method is applied to determine the amount of round bread production at PT Vanessa Bakery. The step involve: determining the fuzzy system, the membership functions, as well as the fuzzy rules. The defuzzification process is applied to determine the amount of total production and to calculate the MAPE value of the Mamdani fuzzy method. The calculated MAPE as much as 5.94%, indicates this method has an excellent forecasting ability because the value is less than 10%. Thus, the Mamdani fuzzy method can be used at PT Vanessa Bakery.

Complex neural fuzzy system and its application on multi-class prediction — A novel approach using complex fuzzy sets, IIM and multi-swarm learning

In this paper, we present a novel complex neural fuzzy approach to multi-class prediction. The complex neural fuzzy system (CNFS) is proposed using complex fuzzy sets (CFSs), fuzzy causalities and multi-swarm machine learning. In general, CFSs are regarded as advanced fuzzy sets with membership degrees defined in the unit disk of the complex plane, in contrast to regular fuzzy sets with membership degrees in the real-valued unit interval [0,1]. The proposed model is composed of the premises designed by CFSs, the consequences designed by Takagi–Sugeno linear functions, and a fuzzy causality layer connecting the premises toward the consequences, and it is able to perform prediction of multiple targets. The usage of fuzzy causality in the proposed model makes difference to traditional fuzzy models using If-Then rules, and gives the freedom and flexibility for model construction. To optimize the proposed CNFS, we present a hybrid learning scheme using the particle swarm optimization with multiple swarms (denoted as PSOmsw) and the Kalman filtering algorithm (denoted as KFA). In the hybrid learning method, the KFA updates the consequence parameters while the PSOmsw evolves the rest parameters of the model. The proposed approach has been tested with experiments using several real-world stock market datasets. Compared with other methods, the proposed approach has shown excellent performance.