Dynamic Evolving Neuro-fuzzy Inference System Research Articles

Rainfall rate estimation over India using global precipitation measurement’s microwave imager datasets and different variants of fuzzy information system

Effective rain rate estimation using satellite-based measurement is imperative for many hydro-meteorological applications. With the recent advancement in satellite products and retrieving algorithms, rain rate estimations are continuously improving. This study provides a comparative performance appraisal of three hybrid machine learning algorithms namely Adaptive Neuro-Fuzzy Inference System (ANFIS), Dynamic Evolving Neuro-Fuzzy Inference System (DENFIS) and Hybrid Fuzzy Inference System (HYFIS) for rain rate estimation using the Global Precipitation Measurement (GPM)’s Microwave Imager (GMI) and ground-based Disdrometer data. The in situ sampling was conducted at four different locations (both land and ocean) across the Indian region and different statistical metrics were used to evaluate the performances of these models. The results showed that HYFIS algorithm has provided better rain rate estimation than ANFIS and DENFIS. The study endorses these neuro-fuzzy models for generating accurate precipitation products and can be considered as an alternative for future satellite retrieval algorithms.

Study for Predicting Land Surface Temperature (LST) Using Landsat Data: A Comparison of Four Algorithms

The soft computing models used for predicting land surface temperature (LST) changes are very useful to evaluate and forecast the rapidly changing climate of the world. In this study, four soft computing techniques, namely, multivariate adaptive regression splines (MARS), wavelet neural network (WNN), adaptive neurofuzzy inference system (ANFIS), and dynamic evolving neurofuzzy inference system (DENFIS), are applied and compared to find the best model that can be used to predict the LST changes of Beijing area. The topographic change is considered in this study to accurately predict LST; furthermore, Landsat 4/5 TM and Landsat 8OLI_TIRS images for four years (1995, 2004, 2010, and 2015) are used to study the LST changes of the research area. The four models are assessed using statistical analysis, coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE) in the training and testing stages, and MARS is used to estimate the important variables that should be considered in the design models. The results show that the LST for the studied area increases by 0.28°C/year due to the urban changes in the study area. In addition, the topographic changes and previously recorded temperature changes have a significant influence on the LST prediction of the study area. Moreover, the results of the models show that the MARS, ANFIS, and DENFIS models can be used to predict the LST of the study area. The ANFIS model showed the highest performances in the training (R2 = 0.99, RMSE = 0.78°C, MAE = 0.55°C) and testing (R2 = 0.99, RMSE = 0.36°C, MAE = 0.16°C) stages; therefore, the ANFIS model can be used to predict the LST changes in the Beijing area. The predicted LST shows that the change in climate and urban area will affect the LST changes of the Beijing area in the future.

Modeling of steelmaking process with effective machine learning techniques

Monitoring and control of the output yield of steel in a steelmaking shop plays a critical role in steel industry. The yield of steel determines how much percentage of hot metal, scrap, and iron ore are being converted into steel ingots. It represents the operational efficiency of the steelmaking shop and is considered as an important performance measure for producing a specific quantity of steel. Due to complexity of the steelmaking process and nonlinear relationship between the process parameters, modeling the input–output process parameters and accurately predicting the output yield in the steelmaking shop is very difficult and has been a major research issue. Statistical models and artificial neural networks (ANN) have been extensively studied by researchers and practitioners to model a variety of complex processes. In the present study, we consider random forests (RF), ANN, dynamic evolving neuro-fuzzy inference system (DENFIS) and support vector regression (SVR) as competitive learning tools to verify the suitability of applications of these approaches and investigate their comparative predictive ability. In the present investigation, 0.00001 of MSE is set as a goal of learning during modeling. Based on real-life data, the computational results depict that the training and testing MSE values of SVR and DENFIS are close to 0.00001 indicating that they have higher prediction ability than ANN and RF. Also, mean absolute percentage prediction errors of the proposed models confirm that the predicted yield based on each method is in good agreement with the testing datasets. Overall, SVR performs best and DENFIS the next best followed by ANN and RF methods respectively. The results suggest that the prediction precision given by SVR can meet the requirement for the actual production of steel.

Prediction of specific cutting energy in natural stone cutting processes using the neuro-fuzzy methodology

Specific cutting energy (SEcut) values are used for the determination of energy requirements of the stone cutting process and are thus useful in predicting the cost and production schedule. In this study, adaptive hybrid intelligence (AHI) techniques were employed to develop SEcut prediction models based on 40 different natural building stones in nineteen different stone processing plants. The feed rate, depth of cut, which are cutting process working parameters, and uniaxial compressive strength, bending strength and point load strength of the rock to be cut which constitute rock physico-mechanical properties were used as the input parameters in the development of SEcut prediction models. The AHI techniques included Adaptive Neuro-Fuzzy Inference System (ANFIS), Dynamic Evolving Neuro-Fuzzy Inference System (DENFIS), and Evolving Fuzzy Neural Networks (EFuNN). Among the AHI techniques, ANFIS gave the best SEcut prediction accuracy. The results also showed that it is possible to predict specific cutting energy of natural stone cutting operations with higher accuracy (R2=0.95) with the developed ANFIS prediction models using depth of cut, feed rate and uniaxial compressive strength values of natural building stones.

Forecasting the NOK/USD Exchange Rate with Machine Learning Techniques

In this paper, we approximate the empirical findings of Papadamou and Markopoulos (2012) on the NOK/USD exchange rate under a Machine Learning (ML) framework. By applying Support Vector Regression (SVR) on a general monetary exchange rate model and a Dynamic Evolving Neuro-Fuzzy Inference System (DENFIS) to extract model structure, we test for the validity of popular monetary exchange rate models. We reach to mixed results since the coefficient sign of interest rate differential is in favor only with the model proposed by Bilson (1978), while the inflation rate differential coefficient sign is approximated by the model of Frankel (1979). By adopting various inflation expectation estimates, our SVR model fits actual data with a small Mean Absolute Percentage Error when an autoregressive approach excluding energy prices is adopted for inflation expectation. Overall, our empirical findings conclude that for a small open petroleum producing country such as Norway, fundamentals possess significant forecasting ability when used in exchange rate forecasting.

Hybrid intelligent systems for predicting software reliability

In this paper, we propose novel recurrent architectures for Genetic Programming (GP) and Group Method of Data Handling (GMDH) to predict software reliability. The effectiveness of the models is compared with that of well-known machine learning techniques viz. Multiple Linear Regression (MLR), Multivariate Adaptive Regression Splines (MARS), Backpropagation Neural Network (BPNN), Counter Propagation Neural Network (CPNN), Dynamic Evolving Neuro-Fuzzy Inference System (DENFIS), TreeNet, GMDH and GP on three datasets taken from literature. Further, we extended our research by developing GP and GMDH based ensemble models to predict software reliability. In the ensemble models, we considered GP and GMDH as constituent models and chose GP, GMDH, BPNN and Average as arbitrators. The results obtained from our experiments indicate that the new recurrent architecture for GP and the ensemble based on GP outperformed all other techniques.

A comparative study of fault density prediction in aspect-oriented systems using MLP, RBF, KNN, RT, DENFIS and SVR models

This paper investigates and empirically evaluates and compares six popular computational intelligence models in the context of fault density prediction in aspect-oriented systems. These models are multi-layer perceptron (MLP), radial basis function (RBF), k-nearest neighbor (KNN), regression tree (RT), dynamic evolving neuro-fuzzy inference system (DENFIS), and support vector regression (SVR). The models were trained and tested, using leave-one-out procedure, on a dataset that consists of twelve aspect-level metrics (explanatory variables) that measure different structural properties of an aspect. It was observed that the DENFIS, SVR, and RT models were more accurate in predicting fault density compared to the MLP, RBF, and KNN models. The MLP model was the worst model, and all the other models were significantly better than it.

NEURO-FUZZY APPROACH TO PREDICTIVE MODELING OF EMISSIONS FROM BIODIESEL POWERED TRANSIT BUSES

The rise in freight passenger transportation is responsible for air pollution, green house gas emissions (especially CO2) and high fuel demand. New engine technology and fuels are discovered and tested throughout the world. Biodiesel, an alternative for diesel, has been seen as a solution. However, the amount of emissions generated by a biodiesel fueled vehicle has not been understood well since most research studies of this kind reported in the literature were conducted in the laboratory. In the present study, emissions (NOx, HC, CO, CO2 and PM) were measured from biodiesel fueled transit buses using an on-road emissions measuring device known as the Portable Emissions Measurement System (PEMS). On-road study is important in terms of understanding the amount of emissions generated under the real traffic and environmental conditions. Emissions were measured on buses fueled with regular diesel (B0), B10 blend (10% biodiesel + 90% diesel) and B20 blend (20% biodiesel + 80% diesel). This paper demonstrates the use of hybrid soft-computing techniques such as the neuro-fuzzy technique for developing emissions prediction models from real-world data. Hybrid soft-computing techniques have been shown to work well in handling data prone to noise and uncertainty, which is characteristic of real-world scenario. Two neuro-fuzzy methodologies were considered in this study: the Adaptive Neuro-Fuzzy Inference System (ANFIS) and the Dynamic Evolving Neuro-Fuzzy Inference System (DENFIS). A brief review of model development, recommended parametric settings, and statistical evaluation of prediction performance of both techniques are discussed. In general, the ANFIS showed better prediction accuracy for the individual emissions compared to DENFIS although the prediction accuracies are comparable.

Knowledge-based evolving connectionist systems for condition evaluation of sustainable roadways: a feasibility study

In place, sustainable rehabilitation of existing deteriorated concrete highways through rubblization is considered to be a green and economical alternative to other options involving total reconstruction, etc. There is currently no standardised method of backcalculating the rubblized pavement layer moduli from pavement non-destructive test data through inverse analysis. This paper explores the feasibility of applying two evolving connectionist systems (ECOS), namely dynamic evolving neuro-fuzzy inference system (DENFIS) and evolving fuzzy neural network (EFuNN), to rubblized pavement moduli backcalculation. It is advantageous to employ ECOS networks for analysing complex engineered systems such as rubblized pavements since ECOS networks are resistant to catastrophic forgetting, have the ability to adapt to and learn new data as soon as they become available, do not have a limit to the amount of knowledge they can store and learn the examples very quickly compared to traditional multi-layered perceptron backpropagation neural networks (MLP-BP NN).

Application of Machine Learning Techniques to Predict Software Reliability

In this paper, the authors employed machine learning techniques, specifically, Back propagation trained neural network (BPNN), Group method of data handling (GMDH), Counter propagation neural network (CPNN), Dynamic evolving neuro–fuzzy inference system (DENFIS), Genetic Programming (GP), TreeNet, statistical multiple linear regression (MLR), and multivariate adaptive regression splines (MARS), to accurately forecast software reliability. Their effectiveness is demonstrated on three datasets taken from literature, where performance is compared in terms of normalized root mean square error (NRMSE) obtained in the test set. From rigorous experiments conducted, it was observed that GP outperformed all techniques in all datasets, with GMDH coming a close second.

Software development cost estimation using wavelet neural networks

Software development has become an essential investment for many organizations. Software engineering practitioners have become more and more concerned about accurately predicting the cost and quality of software product under development. Accurate estimates are desired but no model has proved to be successful at effectively and consistently predicting software development cost. In this paper, we propose the use of wavelet neural network (WNN) to forecast the software development effort. We used two types of WNN with Morlet function and Gaussian function as transfer function and also proposed threshold acceptance training algorithm for wavelet neural network (TAWNN). The effectiveness of the WNN variants is compared with other techniques such as multilayer perceptron (MLP), radial basis function network (RBFN), multiple linear regression (MLR), dynamic evolving neuro-fuzzy inference system (DENFIS) and support vector machine (SVM) in terms of the error measure which is mean magnitude relative error (MMRE) obtained on Canadian financial (CF) dataset and IBM data processing services (IBMDPS) dataset. Based on the experiments conducted, it is observed that the WNN-Morlet for CF dataset and WNN-Gaussian for IBMDPS outperformed all the other techniques. Also, TAWNN outperformed all other techniques except WNN.

Software reliability prediction by soft computing techniques

In this paper, ensemble models are developed to accurately forecast software reliability. Various statistical (multiple linear regression and multivariate adaptive regression splines) and intelligent techniques (backpropagation trained neural network, dynamic evolving neuro–fuzzy inference system and TreeNet) constitute the ensembles presented. Three linear ensembles and one non-linear ensemble are designed and tested. Based on the experiments performed on the software reliability data obtained from literature, it is observed that the non-linear ensemble outperformed all the other ensembles and also the constituent statistical and intelligent techniques.