Search

Wireless sensor networks (WSNs) enable seamless data gathering and communication, facilitating efficient and real-time decision-making in IoT monitoring applications. However, the energy required to maintain communication in WSN-based IoT networks poses significant challenges, such as packet loss, packet drop, and rapid energy depletion. These issues reduce network life and performance, increasing the risk of delayed packet delivery. To address these challenges, this work presents a novel energy-efficient distributed neuro-fuzzy routing model executed in two stages to enhance communication efficiency and energy management in WSN-based IoT applications. In the first stage, nodes with high energy levels are predicted using a fusion of distributed learning with neural networks and fuzzy logic. In the second stage, clustering and routing are performed based on the predicted eligible nodes, incorporating thresholds for energy and distance with two combined metrics. The cluster head (CH) combined metric optimizes cluster head selection, while the next-hop combined metric facilitates efficient multi-hop communication. Extensive simulation results demonstrate that the proposed model significantly enhances network lifetime compared to EANFR, RBFNN T2F, and TTDFP by 9.48%, 25%, and 31.5%, respectively. [ABSTRACT FROM AUTHOR]

Predicting the number of COVID-19 cases offers a reflection of the future, and it is important for the implementation of preventive measures. The numbers of COVID-19 cases are constantly changing on a daily. Adaptive methods are needed for an effective estimation instead of traditional methods. In this study, a novel method based on neuro-fuzzy and FPA is proposed to estimate the number of COVID-19 cases. The antecedent and conclusion parameters of the neuro-fuzzy model are determined by using FPA. In other words, neuro-fuzzy training is carried out with FPA. The number of COVID-19 cases belonging to twenty countries including USA, India, Brazil, Russian, France, UK, Italy, Spain, Argentina, Germany, Colombia, Mexico, Poland, Turkey, Iran, Peru, Ukraine, South Africa, the Netherlands and Indonesia is estimated. Time series is created using the number of COVID-19 cases. Daily, weekly and monthly estimates are realized by utilizing these time series. MSE is used as the error metric. Although it varies according to the example and problem type, the best training error values between 0.000398027 and 0.0286562 are obtained. These best test error values are between 0.0005607 and 0.409867. The best training and test error values are 0.000398027 and 0.0005607, respectively. In addition to FPA, the number of cases is also predicted with the algorithms such as particle swarm optimization, harmony search, bee algorithm, differential evolution and their performances are compared. Success score and ranking are created for all algorithms. The scores of FPA for the daily, weekly and monthly forecast are 71, 77 and 62, respectively. These scores have shown that neuro-fuzzy training based on FPA is successful than other meta-heuristic algorithms for all three prediction types in the short- and medium-term estimation of COVID-19 case numbers. [ABSTRACT FROM AUTHOR]

The Adaptative neuro-fuzzy inference system (ANFIS) has shown great potential in processing practical data from control, prediction, and inference applications, reflecting advantages in both high performance and system interpretability as a result of the hybridization of neural networks and fuzzy systems. Matlab has been a prevalent platform that allows to utilize and deploy ANFIS conveniently. On the other hand, due to the recent popularity of machine learning and deep learning, which are predominantly Python-based, implementations of ANFIS in Python have attracted recent attention. Although there are a few Python-based ANFIS implementations, none of them are directly compatible with scikit-learn, one of the most frequently used libraries in machine learning. As such, this paper proposes Scikit-ANFIS, a novel scikit-learn compatible Python implementation for ANFIS by adopting a uniform format such as fit() and predict() functions to provide the same interface as scikit-learn. Our Scikit-ANFIS is designed in a user-friendly way to not only manually generate a general fuzzy system and train it with the ANFIS method but also to automatically create an ANFIS fuzzy system. We also provide four kinds of representative cases to show that Scikit-ANFIS represents a valuable addition to the scikit-learn compatible Python software that supports ANFIS fuzzy reasoning. Experimental results on four datasets show that our Scikit-ANFIS outperforms recent Python-based implementations while achieving parallel performance to ANFIS in Matlab, a standard implementation officially realized by Matlab, which indicates the performance advantages and application convenience of our software. [ABSTRACT FROM AUTHOR]

Live video broadcasting is a popular application properly considered in lately developed standard, Versatile Video Coding (H.266/VVC). In live video broadcasting, both the bandwidth and buffer volume are limited, while a high quality level is demanded. In order to solve these problems, a Quantization Parameter Estimation Algorithm (QEA) is proposed. The core of the proposed algorithm is a neuro-fuzzy system that changes the Quantization Parameter (QP) gradually to produce a bandwidth-compliant bit rate and prohibit buffer saturation and starvation while providing high quality. The estimation is conducted according to the proportional, integral, and derivative components of the bit error. In other words, the proposed QEA is a Proportional-Integral-Derivative (PID) controller. The optimal parameters of the neuro-fuzzy system are obtained through the training process. The required data set for the training process is established by taking advantage of dynamic programming. The experiments affirm that the proposed approach achieves the target rate with an average error equal to 1.41% and fully respects the buffering boundaries. This method has at least a 2.48% bit rate reduction rather than other QEAs. Meanwhile, the proposed QEA is faster than other algorithms. [ABSTRACT FROM AUTHOR]

Parkinson's Disease (PD) is a complex, degenerative disease that affects nerve cells that are responsible for body movement. Artificial Intelligence (AI) algorithms are widely used to diagnose and track the progression of this disease, which causes symptoms of Parkinson's disease in its early stages, by predicting the results of the Unified Parkinson's Disease Rating Scale (UPDRS). In this study, we aim to develop a method based on the integration of two methods, one complementary to the other, Ensembles of Self-Organizing Map and Neuro-Fuzzy, and an unsupervised learning algorithm. The proposed method relied on the higher effect of the variables resulting from the analysis of the initial readings to obtain a correct and accurate preliminary prediction. We evaluate the developed approach on a PD dataset including speech cues. The process was evaluated with root mean square error (RMSE) and modified R square (modified R2). Our findings reveal that the proposed method is effective in predicting UPDRS outcomes by a combination of speech signals (measures of hoarseness). As the preliminary results during the evaluation showed numbers that proved the worth of the proposed method, such as UPDRS = 0.955 and RMSE approximately 0.2769 during the prediction process. [ABSTRACT FROM AUTHOR]

A neuro-fuzzy computing system combines neural network learning with fuzzy model technique and interpretability into a single unit. Over the past ten years, several neuro-fuzzy systems have been developed. Among these, the adaptive neuro-fuzzy inference system (ANFIS) gives the best design parameters in the least period of time and offers a logical and focused technique for modelling. They are well known for accurately imitating a range of real-world obstacles. Noise pollution's impact on hearing is one such topic that is regularly raised. Specific occupational hazards for laborers include noise, vibration, and low temperatures. Investigating the combined effects of these three physical dangers on employees' physiological markers was the goal of this study, so these three main factors impacting hearing impact are temperature, vibration, and noise intensity, according to the research review. These characteristics interact in complicated, unreliable, and nonlinear ways. Because of this, it is difficult to completely explore using conventional methods. This paper develops a neuro-fuzzy model to predict how noise pollution would affect hearing as a function of noise intensity, vibration, and temperature. On the Fuzzy Logic Toolbox in MATLAB, the model is implemented using ANFIS. The specific fuzzy model that the authors created was employed to assist gather the data for the current investigation. The input/output data sets were split into two groups: 30% were used to assess the model's validity, and the remaining 70% were used to prepare the model for usage. [ABSTRACT FROM AUTHOR]

Nowadays, spoken digit recognition is widely used for various applications. It is a technique used in automatic speech recognition to identify and classify the language spoken by a particular speaker. Several techniques and methods for spoken digit recognition have been proposed. Most of the method includes artificial neural networks, Gaussian mixture models and neuro-fuzzy approach. However, there needs to be a more efficient algorithm that gives more satisfactory and better results. Therefore, to enhance the results in terms of accuracy and running time, a hybrid adaptive neuro-fuzzy approach called KANFIS-SCG is proposed in this paper. This approach combines the Adaptive Network-Based Fuzzy Inference System (ANFIS) and a fast supervised learning algorithm called the Scaled Conjugate Gradient (SCG). In addition, the K-means clustering algorithm is used to reduce the number of fuzzy inference rules. The features are extracted from speech samples using the Mel Frequency Cepstral Coefficient (MFCC) algorithm. Furthermore, the covariance matrix provides more precise information about the features. The experiment was carried out in two languages, namely Hindi and English, and the database was prepared for both languages where utterances were digits from zero to nine. The proposed method KANFIS-SCG gives average recognition rates of 99.12% and 98.54% for Hindi and English, respectively, for the speaker-dependent system, which is better than the existing ANFIS method. [ABSTRACT FROM AUTHOR]

In recent times, the neural image captioning (NIC) is considered as a primitive problem artificial intelligence (AI) in which creates a connection between computer vision (CV) and natural language processing (NLP). However, recent attribute-based and textual semantic attention-based models in NIC still encounter challenges related to irrelevant concentration of the designed attention mechanism on the relationship between extracted visual features and textual representations of corresponding image's caption. Moreover, recent NIC-based models also suffer from the uncertainties and noises of extracted visual latent features from images which sometime leads to the disruption of the given image captioning model to sufficiently attend on the correct visual concepts. To solve these challenges, in this paper, we proposed an end-to-end integrated deep fuzzy-neural network with the unified attention-based semantic-enhanced vision-language approach, called as FuzzSemNIC. To alleviate noises and ambiguities from the extracted visual features, we apply a fused deep fuzzy-based neural network architecture to effectively learn and generate the visual representations of images. Then, the learnt fuzzy-based visual embedding vectors are combined with selective attributes/concepts of images via a recurrent neural network (RNN) architecture to incorporate the fused latent visual features into captioning task. Finally, the fused visual representations are integrated with a unified vision-language encoder–decoder for handling caption generation task. Extensive experiments in benchmark NIC-based datasets demonstrate the effectiveness of our proposed FuzzSemNIC model. [ABSTRACT FROM AUTHOR]

Nowadays, an enormous amount of data is available on the WWW and exponentially growing as well. A lot of users use social networking websites such as Twitter, Facebook, Instagram, and Google+ as common platforms for sharing and exchanging views and opinions on any topics/events. The researchers have considered the reviews and views of the users on these platforms for sentiment analysis, opinion mining, question answering, text summarization, etc. The paper proposes a novel approach for identifying reviews or opinion of users having sarcasm in the text patterns at the clause level. The sentences are classified into four categories such as Simple Sentences, Compound Sentences, Complex Sentences, and Compound-Complex Sentences based on the rules derived from a decision tree. The Simple Sentences and Complex Sentences alone are considered for analysing the sentence patterns where a positive sentiment contrasts with a negative polarity and vice-versa. The decision tree and neuro-fuzzy rules are used on sentence structures to classify the sentences into sarcastic and non-sarcastic sentence patterns. Membership functions are used to map the fuzzy rules and linguistic grading is used for grading the sarcastic patterns. The proposed approach is evaluated on Twitter Dataset and found that the results are promising with the recent and relevant work. [ABSTRACT FROM AUTHOR]

Biochemical oxygen demand (BOD) is one of the most important parameters used for water quality assessment. Alternative methods are essential for accurately prediction of this parameter because the traditional method in predicting the BOD is time-consuming and it is inaccurate due to inconstancies in microbial multiplicity. In this study, the applicability of four hybrid neuro-fuzzy (ANFIS) methods, ANFIS with genetic algorithm (GA), ANFIS with particle swarm optimization (PSO), ANFIS with sine cosine algorithm (SCA), and ANFIS with marine predators algorithm (MPA), was investigated in predicting BOD using distinct input combinations such as potential of hydrogen (pH), dissolved oxygen (DO), electrical conductivity (EC), water temperature (WT), suspended solids (SS), chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (T-P) acquired from two river stations, Gongreung and Gyeongan, South Korea. The applicability of multi-variate adaptive regression spline (MARS) in determination of the best input combination was examined. The ANFIS-MPA was found to be the best model with the lowest root mean square error and mean absolute error and the highest determination coefficient. It improved the root mean square error of ANFIS-PSO, ANFIS-GA, and ANFIS-SCA models by 13.8%, 12.1%, and 6.3% for Gongreung Station and by 33%, 25%, and 6.3% for Gyeongan Station in the test stage, respectively. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) has led to the deployment of many battery-powered sensors in various applications to gather, process, and analyze meaningful data. Clusters of sensors provide for more efficient data collection and increased scalability in such contexts. A low-latency, long-lived routing strategy is described for WSNs that can connect to the Internet of Things. In this research, we present a neuro-fuzzy approach to energy-efficient routing (NFEER) for IoT-enabled WSNs. The novelty of the proposed algorithms is the multiple parameters for the routing in IoT-enabled WSN as consideration of CH distance to sink, cluster size, and residual energy of CH. These variables are used to find the most efficient path across the network, which will help mitigate the hotspot issue. During the operation on the condition "consider only those nodes which have energy greater than the pre-defined threshold energy," the NFEER relies on energy thresholds to restrict the set of candidate nodes. Extensive simulations are performed to specify the effectiveness of the NFEER, and it elongates stability period by 27.98%, 13.97%, and 10.91% as compared to existing protocols. The stability duration, residual energy, network lifetime, and throughput are enhanced by the proposed method as compared to PSO-Kmean, BMHGA, and FSO-PSO. [ABSTRACT FROM AUTHOR]

In this paper, a video Rate Controller (RC) is proposed for the Versatile Video Coding (H.266/VVC) standard. This RC modifies the Quantization Parameter (QP) to keep the buffer state stable and produce the bandwidth-compliant rate simultaneously. It conforms with the specifications of the Variable Bit-Rate (VBR) applications. Intra-frames have high bit-production, which leads to a sudden increase in the rate. Hence, the Group-of-Pictures (GOPs) are categorized into intra-included and intra-excluded. For each type, a distinct neuro-fuzzy QP modifier is designed. The parameters of the neuro-fuzzy QP modifiers are achievable through the training process. However, two data sets are required to train each of the modifiers. These data sets are constructed by the data acquired from the experiments and taking advantage of the q-learning algorithm. Experiments show that the proposed RC achieves the target rate and maintains the buffer state stable. The rate-distortion analysis reveals that the proposed RC has a 2.72% rate-saving capability compared with the one implemented in the reference software of the H.266/VVC. In the case of 4 K sequences, the proposed method saves 5.15% of the rate compared to the rate control algorithm implemented in the reference software. [ABSTRACT FROM AUTHOR]

Forest fire is disastrous to civilizations due to damage to life and property. Forest fire results imbalance of the ecosystem loss of human life and wild animals. Early detection of fire is one of the ways to mitigate this problem. This article proposes a Multimodal framework to identify the fire-prone area of the forest. In this approach, the forest area is divided into different zones. In each zone, two types of sensors are deployed. One type of sensor senses the temperature, relative humidity, drought condition of that zone. Another one is the camera sensors that capture images of that zone simultaneously. All the sensors send the sensed data and image data to the base station. Base station predicts the status (High Active/ Medium Active/Low Active) of the forest zone applying the proposed Multimodal forest fire detection framework. This framework is the integration of the Neuro-fuzzy classification based Sensor model and CNN based Image model. From performance analysis, it is observed that the fire detection accuracy of this proposed Multimodal model is high compared to the individual Sensor and Image model. This model assists the base station in taking necessary action to mitigate fire at that zone in the forest. [ABSTRACT FROM AUTHOR]

To understand the decomposition behavior of biomass oxidation, there is need to analyze the process by some models. An e-tracking system was established and modeled by the object-oriented methodology of kinetic parameters of biomass oxidation. The system calculates kinetic parameters of the biomass oxidation through neuro-fuzzy methodology which is the main core of the e-tracking system. The main attempt in this study was to develop a model for tracking of the biomass oxidation based on different input features. Activation energy and reaction order were the kinetic parameters of the biomass oxidation which were used as the output parameters. Fixed carbon and ash are the most influential factors for the activation energy and reaction order respectively. Oxygen concertation has the smallest impact on the activation energy and reaction order. Designed e-tracking system could have potential for practical applications since it could be updated with more input parameters. [ABSTRACT FROM AUTHOR]

The spread of Covid-19 misinformation on social media had significant real-world consequences, and it raised fears among internet users since the pandemic has begun. Researchers from all over the world have shown an interest in developing deception classification methods to reduce the issue. Despite numerous obstacles that can thwart the efforts, the researchers aim to create an automated, stable, accurate, and effective mechanism for misinformation classification. In this paper, a systematic literature review is conducted to analyse the state-of-the-art related to the classification of misinformation on social media. IEEE Xplore, SpringerLink, ScienceDirect, Scopus, Taylor & Francis, Wiley, Google Scholar are used as databases to find relevant papers since 2018–2021. Firstly, the study begins by reviewing the history of the issues surrounding Covid-19 misinformation and its effects on social media users. Secondly, various neuro-fuzzy and neural network classification methods are identified. Thirdly, the strength, limitations, and challenges of neuro-fuzzy and neural network approaches are verified for the classification misinformation specially in case of Covid-19. Finally, the most efficient hybrid method of neuro-fuzzy and neural networks in terms of performance accuracy is discovered. This study is wrapped up by suggesting a hybrid ANFIS-DNN model for improving Covid-19 misinformation classification. The results of this study can be served as a roadmap for future research on misinformation classification. [ABSTRACT FROM AUTHOR]

Neural network is one among the foremost powerful predictive analytics techniques. This paper considers a wide range of neural network topologies, followed by an examination of how to optimize and analyze neural networks using agriculture data. A back-propagation model was proposed here to create a predictive inference model, with five hidden layers and activation function as sigmoid. Four Kaggle datasets are used here to assess the ability of the proposed neural network approach regarding prediction, and three datasets are gathered from Kerala Agriculture University in real time. Finally, as an optimization strategy, adaptive momentum estimation was utilized, and the applied findings demonstrate that Adam optimizer achieves good prediction accuracy, and F1 score with the help of neuro-fuzzy system and therefore better convergence rate. The proposed method enables more precise control of the direction and step size for updating weight vectors, leading to significantly improved generalization performance. Experimental results reveal that the proposed method shows an accuracy of 97% in average. The suggested approach outperforms other benchmark algorithms such as C4.5, ID3, CART, Naïve Bayes, fuzzy, MLP and hence it can be concluded that the proposed hybridization of back-propagation neural networks with adaptive optimization method clearly verifies the efficiency of Adam optimizer with back propagation for text data. [ABSTRACT FROM AUTHOR]

This article presents a neuro-fuzzy model to predict the resilience index of reinforced concrete moment frames of school buildings. For this purpose, several parameters, including the total height to total width ratio, period, spectral acceleration, peak ground velocity, the effective duration of the earthquake, and distance from the fault, have been considered. The database required to create the neuro-fuzzy model was gathered using the results of the nonlinear analysis of the frames. Based on this analytical data, an approach was introduced and formulated to predict resilience. Further, an existing school building was used to validate the proposed neuro-fuzzy model, and the values were compared. The results indicate a high level of accuracy in calculating the resilience index of the moment frames. In the available methods, it is necessary to perform a large number of time-consuming analyses and complex calculations to measure resilience. However, by using the proposed model, the resilience index could be determined without such analysis. [ABSTRACT FROM AUTHOR]

River water quality is a function of various bio-physicochemical parameters which can be aggregated for calculating the Water Quality Index (WQI). However, it is challenging to model the nonlinearity and uncertain behavior of these parameters. When data is deficient and noisy, it creates missing and conflicting parameters within their complex inter-relationships. It is also essential to model how climatic variations and river discharge affect water quality. The present study proposes a cloud-based efficient and resourceful machine learning (ML) modeling framework using an artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), and advanced particle swarm optimization (PSO). The framework assesses the sensitivity of five critical water quality parameters namely biochemical oxygen demand (BOD), dissolved oxygen (DO), pH, temperature, and total coliform toward WQI of the River Ganges in India. Monthly datasets of these parameters, river flow, and climate components (rainfall and temperature) for a nine-year (2011–2019) period have been used to build the models. We also propose collecting the data by placing various monitoring sensors in the river and sending the data to the cloud for analysis. This helps in continuous monitoring and analysis. Results indicate that ANN and ANFIS capture the nonlinearity in the relationship among water quality parameters with a root mean square error (RMSE) of 7.5 × 10−7 (0.002%) and 1.02 × 10−5 (0.029%), respectively, while the combined ANN-PSO model gives normalized mean square error (NMSE) of 0.0024. The study demonstrates the role of cloud-based machine learning in developing watershed protection and restoration strategies by analyzing the sensitivity of individual water quality parameters while predicting water quality under changing climate and river discharge. [ABSTRACT FROM AUTHOR]

Fog computing has emerged as one of the most important Internet infrastructures for improving service quality, particularly in real-time applications. Due to the convergence in technologies, the scope of the Internet of things (IoT) has evolved to a new dimension, it expands from data collection to device interconnections, and to pre-processing. This acceleration involves cloud and fog computing layers into the system which plays an integral role in IoT data storage and computing. Due to the diversity present in IoT devices, selection of computation devices and allocation of resources are major challenges to be addressed for efficient utilization of resources. In this paper, we presented the offloading and resource allocation model to address the solution to the above challenge. Firstly, a 5-layered neuro-fuzzy model is introduced to retrieve the fuzzy sets and rules which further passes to the fuzzy inference system to model an orchestration decision system. Additionally, to improve the system performance, we have presented the modified least loaded resource allocation algorithm which is adaptively required to reduce the failure rate of the applications. To showcase the efficacy of the model, 4 healthcare applications (augmented reality, patient pre-monitoring, record analysis, and billing systems) are evaluated with their heterogeneous parameters. The simulation findings show that our suggested model improves system performance by lowering network latency by 2.23–9.68 %, computation delay by 3.40–13.66 %, and system performance by 1.03–11.55%. The simulation results demonstrated the suggested model's resilience in terms of network latency, computation time, and failure rate. [ABSTRACT FROM AUTHOR]

Software evolution, an integral part of the software development process, encompasses frequent and numerous changes and updates that may lead to complex and poor-quality systems. Software engineers and quality practitioners continually refactor the software components to mitigate the negative effects of code and design smells. Although these smells are not directly connected to design and coding bugs, they are high indicators of technical debt that may arise as bugs in future software releases. Therefore, refactoring plays a significant role in the overall software evolution process. To fix the underlying code and design smells, they must be first detected and classified. As code smells have been the focus of research in the literature, this paper focuses on detecting design smells. A novel design smell detection scheme is proposed using adaptive neuro-fuzzy approaches. This scheme can be seamlessly integrated with any refactoring scheduling and prioritization models as it is efficiently designed using a fast-training scheme based on a neuro-fuzzy architecture. In addition, the design smell detection solution relies on a reduced set of software metrics. A large custom dataset with more than 30,000 class records is created to assess the performance of the design smell detection solution. The reported performance scores confirm the superiority of the proposed technique over the existing ones. The accuracy, precision, recall, and area-under-the-curve (AUC) scores attained 0.97, 0.98, 0.98, and 0.99, respectively. Thanks to the high detection scores, design smell refactoring sequencing and prioritization routines can be further enhanced. [ABSTRACT FROM AUTHOR]

Blasting is a common technique for rock breakage in the numerous civil and mining engineering activities such as excavation, leveling, and tunneling. However, this technique has several environmental issues, such as ground vibration. More importantly, the peak particle velocity (PPV), as the main indicator of ground vibration, should be considered by engineers/designers due to its potential risk of damaging structures in the nearby area. This research introduces a different modeling procedure to predict PPV resulting from blasting by developing a group method of data handling (GMDH) technique. Data collection and preparation were conducted based on 117 blasting operations at a quarry site, and the effective parameters were considered for prediction purposes. Then, various strategies were defined based on the most important PPV factors, and these strategies were modeled using a variety of parametric studies. After an evaluation process, the best GMDH model was selected for each strategy. As a result, the best GMDH model was related to strategy 3 where four input parameters, i.e., powder factor, charge per delay, sub-drilling, and distance were selected and used. Coefficient correlation results of 0.933 and 0.942 were obtained, respectively, for the training and testing stages of the best GMDH model. To indicate the capability and power of the GMDH model in predicting PPV values, a neuro-fuzzy technique was also proposed for PPV prediction. The obtained results confirmed the power of the developed GMDH model as well as the practical application of this technique in predicting PPV values resulting from blasting. [ABSTRACT FROM AUTHOR]

Purpose: The present study was done to evaluate efficiency of an ensemble learning structure for automatic keratoconus diagnosis and to categorize eyes into four different groups based on a combination of 19 parameters obtained from Pentacam measurements. Methods: Pentacam data from 450 eyes were enrolled in the study. Eyes were separated into training, validation, and testing sets. An ensemble system was used to analyze corneal measurements and categorize the eyes into four groups. The ensemble system was trained to consider indices from both anterior and posterior corneal surfaces. Efficiency of the ensemble system was evaluated and compared in each group. Results: The best accuracy was achieved by the ensemble system with both multilayer perceptron and neuro-fuzzy system classifiers alongside the Naïve Bayes combination method. The accuracy achieved in KC versus N distinction task was equal to 98.2% with 99.1% of sensitivity and 96.2% of specificity for KC detection. The global accuracy was equal to 98.2% for classification of 4 groups, with an average sensitivity of 98.5% and specificity of 99.4%. Conclusion: In this study, authority of an ensemble learning system to work out intricate problems was presented. Despite using fewer parameters, herein, comparable or, in some cases, better results were obtained than methods reported in the literature. The proposed method demonstrated very good accuracy in discriminating between normal eyes and different stages of keratoconus eyes. In some cases, it was not possible to directly compare our results with the literature, due to differences in definitions of KC group as well as differences in selection of items and parameters. [ABSTRACT FROM AUTHOR]

The development of efficient algorithms that perform qualitative operations in the processing and segmentation of graphical images requires good knowledge on both technical and biological aspects of the vision. Due to the subjective nature of the human vision, the evaluation of color identity and closeness, that has been widely used in a variety of image processing problems, is still subjected to research and improvement. This paper analyzes the problem of color tolerance, overviewing past and existing solutions, and suggests a simple Adaptive Neuro-Fuzzy Inference System (ANFIS)-based model as an alternative to the formula-based closeness evaluation. The aim of this research was to develop an alternative to the unified closeness formula, which could be built on expert knowledge or be adapted to fit a particular target. Methods of sample selection, training approaches, and configuration of the ANFIS network were used and a comparative analysis between the ANFIS models and the neural networks was provided. Additionally, the efficiency of regular back-propagation, hybrid, and stochastic learning methods were evaluated along with a critical analysis on the widely used hybrid training method. Experiments were designed and ran on a Java-based platform-independent system, developed as part of our research, to provide flexibility in the modeling and integration for the ANFIS based tasks. [ABSTRACT FROM AUTHOR]

In our current digital life, social networks play a vital role to share the data. A group of hazards that related to the privacy parameters settings of the universal online social network like facebook. The settings corresponding to the user's shared data as well as the data sharing with privacy. A middleware server can split the users with regular social relationships within a same group. The community detection methodology normally request for the full access to the detailed social communications within the users. It is very sensitive to share the personal information like personal images which cause the uncertainty in case of privacy. In this paper, we suggest an Adaptive Framework for privacy preserving in Online Social Networks. It also affords the flexibility in which a third-party server can frequently choose the related sub graph. The emotional detection is constructed with the neuro-fuzzy technique that the fundamental emotions are identified through the singular value decomposition and produces the output of 22 emotions. The experimental results provide the real-time dataset that demonstrate the non-aggregated analysis and practical suggestions highlight the interferences to encourage the secured online social network usage. [ABSTRACT FROM AUTHOR]

Software quality has become a key aspect of any electronic system. In this respect, software reliability is an important quality characteristic and there are many models that aim to estimate the reliability from different perspectives. However, there are no industry established reliability models. There is need to estimate which reliability model has the best performance. In this study several reliability models are analyzed by a soft computing approach, called adaptive neuro-fuzzy inference system (neuro-fuzzy), in order to estimate the models' capability based on root mean square errors (RMSE). Various aspects of accuracy of some of the well-known software reliability models have been used in this work. According to the results Non-Homogeneous Poisson Process Model (NHPP) is the best software reliability model. A combination of Linear Littlewood-Verall (LV) and NHPP is the optimal combination of two software reliability models. In other words, the best results could be achieved if one combines the LV and NHPP models. [ABSTRACT FROM AUTHOR]

This work proposes sentiment analysis for low-resource languages like Hindi using Neuro-Fuzzy Technique. Low-resource languages suffer from the scarcity of resources; consequently, we propose a method that can be implemented for any language. We use information theory for establishing a relation between terms that exists in a sentence. This work proposes a novel approach for calculating feature values using Kullback-Leibler (KL) divergence method. The feature values are employed to calculate the membership values associated with the Fuzzy logic in Neuro-Fuzzy Technique. The novelty of this method lies in its predictive nature that can mitigate the impact generated from un-labeled, unknown data or multi-domain data. We have seen the results for multi-domain data in our experiments. We evaluate our results using Accuracy, Precision, Recall and F1-Score. Our experiments show the efficacy of the proposed approach. It achieved 93.01% accuracy for English dataset and 91.18% accuracy for Hindi dataset which is more than the other state-of-art techniques like Naïve Bayes and SVM. Additionally, we found that our approach provides satisfactory results with multi-domain data as both the datasets were of different domains. [ABSTRACT FROM AUTHOR]

In the data security context, anomaly detection is a branch of intrusion detection that can detect emerging intrusions and security attacks. A number of anomaly detection systems (ADSs) have been proposed in the literature that using various algorithms and techniques try to detect the intrusions and anomalies. This paper focuses on the ADS schemes which have applied fuzzy logic in combination with other machine learning and data mining techniques to deal with the inherent uncertainty in the intrusion detection process. For this purpose, it first presents the key knowledge about intrusion detection systems and then classifies the fuzzy ADS approaches regarding their utilized fuzzy algorithm. Afterward, it summarizes their major contributions and illuminates their advantages and limitations. Finally, concluding issues and directions for future researches in the fuzzy ADS context are highlighted. [ABSTRACT FROM AUTHOR]

This study aims to establish a fatigue life predictive model based on multifractality of road excitations using neuro-fuzzy method to assess the durability of suspension spring. Traditional durability analysis in time domain is complicated and time-consuming due to the needs of large data amount. Thus, it is an idea to adopt an adaptive neuro-fuzzy inference system (ANFIS) for relating the performance of coil spring to the multifractal properties of road excitations, giving a meaningful fatigue life prediction. Different membership function numbers were tested to obtain the optimum membership function number. During the data training process, the checking data was used to test the trained model each Epoch of training for overfitting detection. As a result, the Morrow-based fatigue life prediction model was found to give the most suitable result with three membership functions. The SWT-based model needed five membership functions due to nonlinear properties in the SWT-based fatigue life data. Training process of Morrow-based-ANFIS was stopped at Epoch 8 given its lowest checking root-mean-square-error of 0.6953. SWT-based model recorded a higher error of 0.7940. The neuro-fuzzy models gave accurate fatigue life predictions with 96 % of the data distributed within the acceptance boundary, hence, contributing to an acceptable assessment of coil spring fatigue life based on load multifractality. This study had shown a nonlinear relationship between road multifractality and durability performance of coil spring. Multifractality had been proven an important feature to characterise various road excitations for durability prediction. [ABSTRACT FROM AUTHOR]

This study aims to estimate the depth of anesthesia (DOA) at a safe and appropriate level taking into account the patient characteristics during the induction phase. Bi-spectral Index signal (BIS) as a common approach of controlling DOA generates noise and delays in the initial phase of induction. This may lead to useless information in the process of controlling. Moreover, using the BIS index entails a time-consuming process, high equipping costs, and a lack of accessibility to device accessories. To overcome these problems, we propose a new model of controlling DOA with no need for the use of such an index. Hence, an estimation strategy for DOA is developed applying a feedforward neural network and an adaptive neuro-fuzzy inference estimation model. This model estimates the dose of intravenous anesthetic drugs concerning the patients' needs resulting in optimal drug dose and stable anesthesia depth. The proposed estimations are tested by sensitivity analysis being compared with real data obtained from the classical model (PK-PD) revised approach and BIS approach on 13 patients undergoing surgery. The results show an accuracy of 0.999, indicative of a high-validated model. Compared to BIS, our proposed model not only controls DOA accurately but also achieves outcomes in practice successfully. Some practical implications for future research and clinical practice are also suggested. [ABSTRACT FROM AUTHOR]

Virtualization provides the flexibility to distribute the workload among physical servers to reduce overall electrical power consumption, through the consolidation of Virtual Machines (VMs). Many research projects have been done on VM migration to reduce energy consumption in data centers while ensuring a high level of adherence to the Service Level Agreements (SLA). Service levels of running applications are likely to be negatively affected during a live VM migration. For this reason, in this paper, we propose a new intelligent VM migration approach, called CLANFIC, which utilizes modified Cellular Learning Automata based Evolutionary Computing (CLA-EC) and neuro-fuzzy to minimize the number of VM migrations and improve energy consumption. This goal is achieved by utilizing an optimized placement method and delaying migration time based on future resource demand prediction. This algorithm reduces the number of migrations in two steps (i) finding the optimal virtual machine placement and replacement on physical servers by using modified CLA-EC (ii) predicting future resource usage in each host by a neuro-fuzzy algorithm to prevent unnecessary migrations. The experimental results on the real workload traces from PlanetLab show that the proposed method reduces the mean migration number, energy consumption, and SLA violation of the data center by 59.05%, 8.5%, and 70.76%, respectively. [ABSTRACT FROM AUTHOR]

The poor management of hotspots in electrical systems often leads to very devastating consequences; an extremity is fire outbreak which could result in loss of lives and/or properties. Hence, the necessity to effectively manage or handle electrical hotspots cannot be overstated. This paper proposes a model for arresting the occurrence of the destructive activities of electrical hotspots in industries by controlling the temperature, humidity and dust density within electrical components/equipment. To this end, a thermal imaging camera is employed for the detection of various locations and magnitudes of hotspots within electrical systems. Based on the globally approved industrial standards for the prevention of thermally induced electrical systems failure, each of the electrical components and equipment [whose thermal excursion is beyond the allowable temperature rise under measured load values (i.e. ΔTcorr)] is identified and treated by adopting the recommended actions. Additionally, a fuzzy logic control (FLC) system is designed. This is further developed into an adaptive neuro-fuzzy inference system (ANFIS) for the control of the operation of the air handling unit (AHU) and the aspirator suction speed. This arrangement, thus, leads to heat reduction and dust elimination within the electrical components/equipment in industrial space, thus preventing the destructive effects of the occurrence of hotspots. However, for the sake of graphical representations of this scheme, the MATLAB environment is created for the generation of the optimum temperatures at various locations within the electrical systems. From this development, it is established that the framework has very high potential to eliminate the catastrophic effects of hotspots in electrical systems. [ABSTRACT FROM AUTHOR]

The segmentation of the lesion plays a core role in diagnosis and monitoring of multiple sclerosis (MS). Magnetic resonance imaging (MRI) is the most frequent image modality used to evaluate such lesions. Because of the massive amount of data, manual segmentation cannot be achieved within a sensible time that restricts the usage of accurate quantitative measurement in clinical practice. Therefore, the need for effective automated segmentation techniques is critical. However, a large spatial variability between the structure of brain lesions makes it more challenging. Recently, convolutional neural network (CNN), in particular, the region-based CNN (R-CNN), have attained tremendous progress within the field of object recognition because of its ability to learn and represent features. CNN has proven a last-breaking performance in various fields, such as object recognition, and has also gained more attention in brain imaging, especially in tissue and brain segmentation. In this paper, an automated technique for MS lesion segmentation is proposed, which is built on a 3D patch-wise R-CNN. The proposed system includes two stages: first, segmenting MS lesions in T2-w and FLAIR sequences using R-CNN, then an adaptive neuro-fuzzy inference system (ANFIS) is applied to fuse the results of the two modalities. To evaluate the performance of the proposed method, the public MICCAI2008 MS challenge dataset is employed to segment MS lesions. The experimental results show competitive results of the proposed method compared with the state-of-the-art MS lesion segmentation methods with an average total score of 83.25 and an average sensitivity of 61.8% on the MICCAI2008 testing set. [ABSTRACT FROM AUTHOR]

Spinal posture including thorax/pelvis orientations as well as loads on the intervertebral discs are crucial parameters in biomechanical models and ergonomics to evaluate the risk of low back injury. In vivo measurement of spinal posture toward estimation of spine loads requires the common motion capture techniques and laboratory instruments that are costly and time-consuming. Hence, a closed loop algorithm including an artificial neural network (ANN) and fuzzy logic is proposed here to predict the L5–S1 segment loads and thorax/pelvis orientations in various 3D reaching activities. Two parts namely a fuzzy logic strategy and an ANN from this algorithm; the former, developed based on the measured postures of 20 individuals, is to determine 3D orientations of the thorax/pelvis during the various activities while the latter, developed based on the predicted L5–S1 loads by a detailed musculoskeletal model of the spine, is to estimate compression/shear forces at the L5–S1 disc. The fuzzy logic rules are extracted based on Sugeno inference engine and the ANN is trained by LevenbergMarquardt algorithm. To evaluate the performance of the proposed strategy, the comparison between the predicted values, the target values and the presented values in the literature are reviewed. The comparison demonstrated that the proposed algorithm had a promising performance. The maximum relative error for all predictions was ~19 % and with respect to the target values while this error for the literature's values was ~37 %. Also, the average improvement of the proposed strategy was ~17 % with respect to the presented strategy in the literature. [ABSTRACT FROM AUTHOR]

This paper concerns the application of a neuro-fuzzy learning method based on data streams for high impedance fault (HIF) detection in medium-voltage power lines. A wavelet-packet-transform-based feature extraction method combined with a variation of evolving neuro-fuzzy network with fluctuating thresholds is considered for recognition of spatial–temporal patterns in the data. Wavelet families such as Haar, Symlet, Daubechie, Coiflet and Biorthogonal were investigated as a way to provide the most discriminative features for fault detection. The proposed evolving neuro-fuzzy classification model has shown to be particularly suitable for the problem because the HIF environment is subject to concept changes. Different from other statistical and intelligent approaches to the problem, the developed neuro-fuzzy model for HIF classification is not only parametrically, but also structurally adaptive to cope with nonstationarities and novelties. New neurons and connections are incrementally added to the neuro-fuzzy network when necessary for the identification of new patterns, such as faults and usual transients including sag, swell and spikes due to the switching of 3-phase capacitors and energization of transformers. Experimental evaluations compare the proposed classifier with other well-established computational intelligence methods, viz. multilayer perceptron neural network, learning vector quantization neural network and a support vector machine model. Results have shown that the evolving neuro-fuzzy system is effective and robust to changes. The system is able to maintain its detection and classification accuracy even in situations in which other classifiers exhibit a significant drop in accuracy due to gradual and abrupt changes of the fault patterns. Fuzzy rules are useful for interpretability purposes and help to enhance model credibility for decision making. [ABSTRACT FROM AUTHOR]

ANFIS is an artificial intelligence technique which is composed of a combination of artificial neural networks and fuzzy inference system. Due to its structure, it is used in modeling and identifying numerous systems in various fields. The training process of ANFIS is important to obtain effective results with it. So, a successful training algorithm should be used. In this study, the ANFIS is trained by using ABC algorithm for the solution of the real-world problem. For this purpose, the number of foreign visitors coming to Turkey from the USA, Germany, Bulgaria, France, Georgia, the Netherlands, England, Iran and Russia is estimated. In addition, total number of visitors coming to Turkey is also predicted. In applications, 150 months of data between July 2002 and December 2014 are utilized and a time series is created using these data. The results obtained using ABC algorithm are compared with GA, DE, HS and PSO. As a conclusion, it is seen that the results obtained using ABC algorithm for estimation of number of foreign visitors are more successful than other optimization methods. [ABSTRACT FROM AUTHOR]

Today, researchers face multiple challenges identifying clay mineral types and lithofacies from well log data. This research paper hopes to offer new insight into this particular challenge. Formation evaluation characteristics play a significant role in the exploration and production of future and current oil and gas fields. The proposed methodology in this study uses an integrated approach that includes: (1) numerical equations, (2) Neuro-Fuzzy neural networks, (3) cross plots, and (4) statistical analyses. This proposed integrated approach is capable of dramatically improving the accuracy of the results. Well logging data provide valuable information for identifying lithofacies, clay mineralogy types, as well as other important hydrocarbon reservoir characteristics. Talhar Shale in the Southern Lower Indus Basin, Pakistan, is composed of interbedded shale, sand, and shaly-sand, intervals that have been identified via the lithological interpretation process of well logs. Talhar Shale contains montmorillonite type clay with minor amounts of illite, glauconite, and various micas that can be easily identified by natural gamma ray absorption profiles, as well as through ratio logs, bulk density log, and photoelectric absorption index log. These interpretations can be further confirmed via cross plots and other statistical analyses. This approach consists of a comprehensive study of well logging data and thus can lend itself to be a helpful component in characterizing the hydrocarbon structures of the Talhar Shale. [ABSTRACT FROM AUTHOR]

Image-processing methods are applied in many environments to solve various practical problems. Recently, considerable research has been conducted on analysing retinal images, which show the blood vessels in the human eye, to detect diseases in patients. Various diseases related to the eyes can be located by detecting changes in the retinal blood cells. This paper proposes two analysis methods, namely the fuzzy c-means and neuro-fuzzy methods, which analyse optical coherence tomography images of retinal blood cells. The system was implemented in a MATLAB experimental setup, while the efficiency was evaluated with respect to time and accuracy. The performance analysis of the fuzzy c-means and neuro-fuzzy systems indicates that the neuro-fuzzy system has a higher accuracy (93.16%) and a lower processing time (2 s) than the k-means clustering, expectation maximisation, fuzzy clustering and neural network methods. [ABSTRACT FROM AUTHOR]

Modeling pan evaporation (Epan) estimation is a vital issue in water resources management because it directly affects water reservoir and water supply systems. In the developing countries (e.g., India), Epan data are generally limited, and in such a circumstance, theoretical estimates from available climatic data could be beneficial. The study investigates the capability of three adaptive neuro-fuzzy methods, adaptive neuro-fuzzy inference system (ANFIS)–embedded grid partition (GP), subtractive clustering (SC), and fuzzy c-means clustering (FCM), in estimation of monthly pan evaporation using climatic inputs of minimum and maximum air temperatures, wind speed, sunshine hours, and relative humidity obtained from two stations, Uttarakhand, India. Cross validation method is applied by dividing data into three equal parts, and methods are tested using each part. Methods are evaluated by applying various combinations of inputs and using root mean square error (RMSE), mean absolute error (MAE), Nash–Sutcliffe efficiency (NSE), and determination coefficient (R2) criteria. The ANFIS-FCM is found to be superior to the ANFIS-GP and ANFIS-SC methods in Epan modeling. Cluster-based proposed neuro-fuzzy method increases performance of the best ANFIS-GP and ANFIS-SC models with respect to RMSE by about 9–14% for the both stations. The three ANFIS methods are also compared with each other and Stephen Stewart (SS) method by dividing data into three stages, training, validation, and test. The results indicate the superior accuracy of the ANFIS methods to SS for the same input variables. The ANFIS-FCM generally produces better Epan estimates than the other two ANFIS methods. [ABSTRACT FROM AUTHOR]

In order to broaden the study of the most popular and general Takagi–Sugeno (TS) system, we propose a complex-valued neuro-fuzzy inference system which realises the zero-order TS system in the complex-valued network architecture and develop it. In the complex domain, boundedness and analyticity cannot be achieved together. The splitting strategy is given by computing the gradients of the real-valued error function with respect to the real and the imaginary parts of the weight parameters independently. Specifically, this system has four layers: in the Gaussian layer, the L-dimensional complex-valued input features are mapped to a Q-dimensional real-valued space, and in the output layer, complex-valued weights are employed to project it back to the complex domain. Hence, split-complex valued gradients of the real-valued error function are obtained, forming the split-complex valued neuro-fuzzy (split-CVNF) learning algorithm based on gradient descent. Another contribution of this paper is that the deterministic convergence of the split-CVNF algorithm is analysed. It is proved that the error function is monotone during the training iteration process, and the sum of gradient norms tends to zero. By adding a moderate condition, the weight sequence itself is also proved to be convergent. [ABSTRACT FROM AUTHOR]

Content-centric networking (CCN) is gradually becoming the alternative approach to the traditional Internet architecture through enlightening information (content) distribution on the Internet with content names. The growing rate of Internet traffic has adapted a content-centric architecture to better serve the user requirement of accessing a content. For enhancing content delivery, ubiquitous in-network caching is utilized to store a content in each and every router by the side of the content delivery path. From the study, it is evaluated that a better performance can be achieved when caching is done by a subset of CRs instead of all CRs in a content delivery path. Motivated by this, we proposed an adaptive neuro-fuzzy inference system-based caching (ANFIS-BC) scheme for CCN to improve the cache performance. The proposed ANFIS-BC scheme utilizes the feature of centrality-measures for selection of a router for caching in a network. Our results demonstrated that the ANFIS-BC scheme consistently achieves better caching gain across the multiple network topologies. [ABSTRACT FROM AUTHOR]

Classification of spatial exploration data for exploration targeting using neuro-fuzzy models means that the many spatial values have to be simplified and assigned to a few classes. The simplification of complex geological information, which illustrates a high degree of variability, results in overly simplistic models based on the presumption of homogeneous earth. However, such an assumption is not valid. In this paper, we illustrate the superiority of using continuously weighted spatial evidence values compared to discretely weighted evidence data, and how continuously weighted spatial evidence values can increase the efficiency of neuro-fuzzy exploration targeting models. The results of this study demonstrate that neuro-fuzzy targeting model generated with continuously weighted spatial evidence values is superior to that of the neuro-fuzzy model generated with discretely weighted exploration evidence data. [ABSTRACT FROM AUTHOR]

Sediment rating curves (SRCs) have been recognized as the most popular method for estimating sediment in the hydrology of river sediments and in watersheds. In this regard, in order to compare and correct estimation methods of river sediment load, estimated rates of several univariate types of SRCs and a multivariate type of SRCs (MSRCs) were studied using the neuro-fuzzy and tree regression models in five selective hydrometric stations of different climatic zones of Iran and with various indexes of the accuracy (AI) and the precision (PI). The results of the data analysis showed that the mean of the AI of neuro-fuzzy and tree regression models in selective stations is 151 and 536%, respectively, which shows the low efficiency compared with SRCs. Also according to the results, the best rate of the AI of the MSRCs belongs to the Glink station with the rate of 1.12. Also, the average value of the AI of MSRCs is 1.15 which is an acceptable amount of the other considered various methods. [ABSTRACT FROM AUTHOR]

This paper proposes an optimal hybrid neuro-fuzzy/fuzzy controller based on maximum power point tracking (MPPT) technique and voltage regulation for photovoltaic lead-acid battery charging system through the constant current and constant voltage (CC-CV) charge, denoted by NFC-CC/FLC-CV. The parameter optimization of NFC and FLC, including rule selection, based on multi-objective genetic algorithm (MOGA) is applied to the NFC-CC design to improve the tracking accuracy while reducing complexity. By means of genetic optimization, the number of fuzzy rules can be greatly reduced by 50%. In addition, GA is applied to the FLC-CV design to increase voltage regulation (VR) accuracy. After satisfying the stability condition through the solutions determined by MOGA and GA, the performances of controllers under rapidly-changing weather are evaluated tradeoff by several creteria, including transient response, stabilized accuracy, charging time, and energy utilization and charging efficiency. As results, the proposed controller outperforms the other existing controllers with the fastest rise time without overshoot, the highest MPPT and VR accuracy with negligible oscillations, a 12-23% reduction in charging time, and an increase of 5-15% and 1-6% in energy utilization and charging efficiency. Furthermore, it provides superior results in terms of computational complexity by achieving the minimum number of multiplications and system parameters, and high reliability with the lowest Akaike information criterion (AIC). [ABSTRACT FROM AUTHOR]

Squirrel-cage induction motors are widely used in a number of applications throughout the world. This paper proposes a neuro-fuzzy approach to identify and to classify a typical fault related to the induction motor damage, such as broken rotor bars. Two fuzzy classifiers are obtained by an adaptive-network-based fuzzy inference system model whose parameters can be identified by using the hybrid learning algorithm. A Hall effect sensor was installed between two stator slots of the induction machine, and a magnetic flux density variation is measured according to the failure. The data from the Hall sensor were used to extract some harmonic components by applying fast Fourier transform. Thus, some frequencies and their amplitudes were considered as inputs for the proposed fuzzy model to detect not only adjacent broken bars, but also noncontiguous faulted scenarios. In the present work it is not necessary to estimate the rotor slip, as required by the traditional condition monitoring technique, known as motor current signature analysis. This method was able to detect broken bars for induction motor running at low-load or no-load condition. The intelligent approach was validated using some experimental data from a 7.5-kW squirrel-cage induction machine. [ABSTRACT FROM AUTHOR]

The unmanned aerial vehicles can have complicated dynamics and kinematics that governs the flight of such multirotor devices. PID type controllers are one of the most popular approaches with Raspberry Pi 3 platform for stability of the flight. However, in dynamic environment they are limited in performance and response times. The autonomous tuning of the controller parameters according to the state of the environment with assistance of the adaptive neuro-fuzzy inference system is a well known approach. This paper provides implementation details and feasibility of such a controller with Raspberry Pi 3 platform for use in geological wireless sensing environments. The proposed neuro-fuzzy controller is developed for a Raspberry Pi 3 platform and tested on a physical quadrotor drone and compared to the conventional PID controller during flight. [ABSTRACT FROM AUTHOR]

The neuro-fuzzy controller applies the neural network learning techniques to tune the membership functions and keeps the semantics of the fuzzy logic controller intact. Hence benefits of both the neural network and fuzzy logic controller are taken into consideration. In this study, to predict the bearing capacity of a stone column, application of Adaptive Neuro-fuzzy Inference System (ANFIS) is presented. To train and test the data sets, 105 data pairs are collected from the previous technical literature. These data sets include the data of stone and sand columns. The spacing of the columns varies from 1.5 to 10 times the diameter. The undrained cohesion varies from 7 to 400 kPa. Both experimental and analytical data are included in the collection. To test the trained ANFIS models, data are collected from physical experiments on plate load test and numerical analysis with PLAXIS-2D. For the comparative study, ANFIS models combined with plate load test results and analytical results, three ANFIS models are developed. A comparative study on the accuracy of prediction by these three models is discussed. [ABSTRACT FROM AUTHOR]

This paper investigates the ability of four artificial intelligence techniques, including artificial neural network (ANN), radial basis neural network (RBNN), adaptive neuro-fuzzy inference system (ANFIS) with grid partitioning, and ANFIS with fuzzy c-means clustering, to predict the peak and residual conditions of actively confined concrete. A large experimental test database that consists of 377 axial compression test results of actively confined concrete specimens was assembled from the published literature, and it was used to train, test, and validate the four models proposed in this paper using the mentioned artificial intelligence techniques. The results show that all of the neural network and ANFIS models fit well with the experimental results, and they outperform the conventional models. Among the artificial intelligence models investigated, RBNN model is found to be the most accurate to predict the peak and residual conditions of actively confined concrete. The predictions of each proposed model are subsequently used to study the interdependence of critical parameters and their influence on the behavior of actively confined concrete. [ABSTRACT FROM AUTHOR]

An adaptive iterative learning controller (ILC) is designed for a class of nonlinear discrete-time systems based on data driving control (DDC) scheme and adaptive networks called fuzzy rules emulated network (FREN). The proposed control law is derived by using DDC scheme with a compact form dynamic linearization for iterative systems. The pseudo-partial derivative of linearization model is estimated by the proposed tuning algorithm and FREN established by human knowledge of controlled plants within the format of IF-THEN rules related on input-output data set. An on-line learning algorithm is proposed to compensate unknown nonlinear terms of controlled plant, and the controller allows to change desired trajectories for other iterations. The performance of control scheme is verified by theoretical analysis under reasonable assumptions which can be held for a general class of practical controlled plants. The experimental system is constructed by a commercial DC motor current control to confirm the effectiveness and applicability. The comparison results are addressed with a general ILC scheme based on DDC. [ABSTRACT FROM AUTHOR]

In this research a framework is developed to predict the drinking water quality through the neural network models. A fuzzy rule-based system and similarity measure algorithm yield a water quality index for different sampling locations in a water distribution network (WDN), and a neural network is trained using the quality indices. Different sources of uncertainty exist in this model, including deficient, missing, and noisy data, conflicting water quality parameters and subjective information. Hourly and monthly data from Quebec City WDN are used to illustrate the performance of the proposed neuro-fuzzy model. Also, historical data from 52 sampling locations of Quebec City network is utilized to develop the rule-based model in order to train the neural network. Water quality is evaluated by categorizing quality parameters in two groups including microbial and physicochemical. Two sets of rules are defined using expert knowledge to assign water quality grades to each sampling location in the WDN. The fuzzy inference system outputs are deffuzzified using a similarity measure algorithm in this approach. The fuzzy inference system acts as a decision making agent. A utility function provides microbial and physicochemical water quality indices. Final results are used to train the neural network. In the proposed framework, microbial and physicochemical quality of water are predicted individually. [ABSTRACT FROM AUTHOR]

In this work an Adaptive Neuro-Fuzzy Inference System (ANFIS) was used to model the periodic performance of some multi-input single-output (MISO) processes, namely: brewery operations (case study 1) and soap production (case study 2) processes. Two ANFIS models were developed to model the performance of the two processes under study. The results of the study show that for brewery operations, ANFIS model 2 with a correlation coefficient of 0.9972, as against 0.9956 for ANFIS model 1, had a better correlation than an equivalent MAMDANI fuzzy model. On the order hand, for soap production process, ANFIS model 1 had better correlation with an equivalent MAMDANI model. Generally, there is a general agreement among the models on the periodic performance of the processes. Thus, all the models show that for the brewery, the best performance was in the period 2010-2011 and the period 2008-2009 was the worst. Similarly, for the soap production process, the best performance was in 2011 and the worst in 2012. The results show that a combination of transfer function and ANFIS could be used effectively to model process performance. [ABSTRACT FROM AUTHOR]