Interpretation of artificial neural networks by means of fuzzyrules

This paper presents an extension of the method presented by Benitez et al (1997) for extracting fuzzy rules from an artificial neural network (ANN) that express exactly its behavior. The extraction process provides an interpretation of the ANN in terms of fuzzy rules. The fuzzy rules presented are in accordance with the domain of the input variables. These rules use a new operator in the antecedent. The properties and intuitive meaning of this operator are studied. Next, the role of the biases in the fuzzy rule-based systems is analyzed. Several examples are presented to comment on the obtained fuzzy rule-based systems. Finally, the interpretation of ANNs with two or more hidden layers is also studied.     

Classification of modulation signals using statistical signal characterization and artificial neural networks

Modulation recognition systems have to be able to correctly classify the incoming signal modulation scheme in the presence of noise. A new method for classification of analogue and digital modulated signals at low signal-to-noise ratio (SNR) is introduced in this paper. This method uses the statistical signal characterization (SSC) to extract parameters to classify the different modulation signals. The SSC technique produces a set of four numerical parameters for a specific modulated signal. Subsequent comparison of these parameters to those of other waveforms provides the basis for our classification system. The results of SSC technique are applied to an artificial neural network (ANN) to have a robust classification system in the presence of noise down to SNR of 3 dB. No a priori information is required by this technique about the set of input waveforms. The input to the classification system can be analogue or digital signals or a combination of both. The proposed technique shows a 100% efficiency of classification of analogue signals or digital signals at SNR of 7 dB. This classification efficiency reduces to 83% and 86% for analogue or digital signals at SNR of 3 dB. The SSC technique shows better classification results in comparison with other techniques with an important advantage over other methods, which is the simplicity of the neural network needed with this technique due to the small number of features used in the classification.     

Prediction of building energy consumption by using artificial neural networks

In this study, the main objective is to predict buildings energy needs benefitting from orientation, insulation thickness and transparency ratio by using artificial neural networks. A backpropagation neural network has been preferred and the data have been presented to network by being normalized. The numerical applications were carried out with finite difference approach for brick walls with and without insulation of transient state one-dimensional heat conduction. Three different building samples with different form factors (FF) were selected. For each building samples 0–2.5–5–10–15 cm insulations are assumed to be applied. Orientation angles of the samples varied from 0° to 80° and the transparency ratios were chosen as 15–20–25%. A computer program written in FORTRAN was used for the calculations of energy demand and ANN toolbox of MATLAB is used for predictions. As a conclusion; when the calculated values compared with the outputs of the network, it is proven that ANN gives satisfactory results with deviation of 3.43% and successful prediction rate of 94.8–98.5%.     

Ten steps modeling of electrolysis processes by using neural networks

Neural networks have been developed to model the electrolysis of wastes polluted with phenolic compounds, including phenol, 4-chlorophenol, 2,4-dichlorophenol, 2,4,6-trichlorophenol, 4-nitrophenol and 2,4-dinitrophenol. They enable the prediction the Chemical Oxygen Demand of a treated waste as a function of the initial characteristics (pollutant concentration, pH), operation conditions (temperature, current density) and current charge passed. A consistent set of experimental data was obtained by electrochemical oxidation with conductive diamond electrodes, used to treat synthetic aqueous wastes.Several modeling strategies based on simple and stacked neural networks, with different transfer functions into the hidden and output layers, have been considered to obtain a good accuracy of the model. Global errors during the training stage were under 3% and those of the validation stage were under 4%, demonstrating that the neural network based technique is appropriate for modeling the system.The generalization capability of the neural networks was also tested in realistic conditions where Chemical Oxygen Demand was predicted with errors around 5%. Therefore, the developed neural models can be used in industry to determine the required treatment period, to obtain the discharge limits in batch electrolysis processes, and it is a first step in the development of process control strategies.The ten step methodology was applied to the neural network based process modeling.     

Prediction of gait speed from plantar pressure using artificial neural networks

The goal of this study was to predict gait speed over the entire cycle in reference to plantar pressure data acquired by means of the insole-type plantar pressure measuring device (Novel Pedar-x system). To predict gait speed, the artificial neural network is adopted to develop the model to predict gait speed in the stance phase (Model I) and the model to predict gait speed in the swing phase (Model II). The predicted gait speeds were validated with actual values measured using a motion capturing system (VICON 460 system) through a five-fold cross-validation method, and the correlation coefficients (R) for the gait speed were 0.963 for normal walking, 0.978 for slow walking, and 0.950 for fast walking. The method proposed in this study is expected to be widely used clinically in understanding the progress and clarifying the cause of such diseases as Parkinsonism, strike, diabetes, etc. It is expected that the method suggested in this study will be the basis for the establishment of a new research method for pathologic gait evaluation.     

Biomass retrieval from high-dimensional active/passive remote sensing data by using artificial neural networks

Retrieval of the biomass parameters from active/passive microwave remote sensing data is performed based on an iterative inversion of the artificial neural network (ANN). The ANN is trained by a set of the measurements of active and passive remote sensing and the ground truth data versus Day of Year during growth. Once the ANN training is complete, the ANN can be used to retrieve the temporal variations of the biomass parameters from another set of observation data. The retrieved biomass include canopy height, canopy water content and dry matter fraction, and the wetness of the underlying land. Two examples for wheat and oat are illustrated. The retrieved biomass parameters agree well with the real data of the ground truth.     

Soft-sensing estimation of plant effluent concentrations in a biological wastewater treatment plant using an optimal neural network

Recent studies into the estimation and control of an activated sludge process (ASP) at a wastewater treatment plant suggest that artificial-intelligence methods, such as neural networks, fuzzy systems and genetic algorithms, can improve the plant performance in terms of reduced operation costs and improved effluent quality. In this paper, a neural-network-based soft sensor is developed for the on-line prediction of effluent concentrations in an ASP in terms of primary hard-to-measure variables, such as chemical oxygen demand, total nitrogen content and total suspended solids, starting from secondary on-line easy-to-measure variables, such as oxygen and nitrogen compound concentrations in biological tanks, input flow rate and alkalinity, among others. An algorithm based on principal component analysis is applied to select the optimal net input vectors for the soft sensor, using an appropriated number of samples of the secondary variables set. The proposed soft sensor is tested on the ASP of a large-scale municipal wastewater treatment plant running under the GPS-X simulation frame and validated with operational gathered data. Satisfactory low values for mean and maximum absolute prediction errors are obtained, even when high values of sampling time of primary variables are set, as it is frequently done during monitoring operation. In this way, data-driven soft-sensors based on neural networks can become valuable tools for plant operators for the recognition of operational states in terms of low cost and efficient prediction of primary process variables such as chemical oxygen demand, total nitrogen content and total suspended solids, therefore avoiding the acquisition of expensive and sometimes unreliable instruments for measuring nutrient concentrations in plant.     

Sleep spindle detection using artificial neural networks trained with filtered time-domain EEG: a feasibility study

An artificial neural network (ANN) based on the Multi-Layer Perceptron (MLP) architecture is used for detecting sleep spindles in band-pass filtered electroencephalograms (EEG), without feature extraction. Following optimum classification schemes, the sensitivity of the network ranges from 79.2% to 87.5%, while the false positive rate ranges from 3.8% to 15.5%. Furthermore, due to the operation of the ANN on time-domain EEG data, there is agreement with visual assessment concerning temporal resolution. Specifically, the total inter-spindle interval duration and the total duration of spindles are calculated with 99% and 92% accuracy, respectively. Therefore, the present method may be suitable for investigations of the dynamics among successive inter-spindle intervals, which could provide information on the role of spindles in the sleep process, and for studies of pharmacological effects on sleep structure, as revealed by the modification of total spindle duration.     

Control rules of aeration in a submerged biofilm wastewater treatment process using fuzzy neural networks

The present work is part of a global development of reliable real-time control and supervision tools applied to wastewater pollution removal processes. In these processes, oxygen is a key substrate in animal cell metabolism and its consumption is thus a parameter of great interest for the monitoring. In this paper, an integrated neural-fuzzy process controller was developed to control aeration in an Aerated Submerged Biofilm Wastewater Treatment Process (ASBWTP). In order to improve the fuzzy neural network performance, the self-learning ability embedded in the fuzzy neural network model was emphasized for improving the rule extraction performance. The fuzzy neural network proves to be very effective in modeling the aeration performs better than artificial neural networks (ANN).For comparing between operation with and without the fuzzy neural controller, an aeration unit in an Aerated Submerged Biofilm Wastewater Treatment Process (ASBWTP) was picked up to support the derivation of a solid fuzzy control rule base. It is shown that, using the fuzzy neural controller, in terms of the cost effectiveness, it enables us to save almost 33% of the operation cost during the time period when the controller can be applied. Thus, the fuzzy neural network proved to be a robust and effective DO control tool, easy to integrate in a global monitoring system for cost managing.     

Predicting the type of pregnancy using artificial neural networks and multinomial logistic regression: a comparison study

Some medical and epidemiological surveys have been designed to predict a nominal response variable with several levels. With regard to the type of pregnancy there are four possible states: wanted, unwanted by wife, unwanted by husband and unwanted by couple. In this paper, we have predicted the type of pregnancy, as well as the factors influencing it using two different models and comparing them. Regarding the type of pregnancy with several levels, we developed a multinomial logistic regression and a neural network based on the data and compared their results using three statistical indices: sensitivity, specificity and kappa coefficient. Based on these three indices, neural network proved to be a better fit for prediction on data in comparison to multinomial logistic regression. When the relations among variables are complex, one can use neural networks instead of multinomial logistic regression to predict the nominal response variables with several levels in order to gain more accurate predictions.     

Modelling of a vertical ground coupled heat pump system by using artificial neural networks

This paper describes the applicability of artificial neural networks (ANNs) to estimate of performance of a vertical ground coupled heat pump (VGCHP) system used for cooling and heating purposes experimentally. The system involved three heat exchangers in the different depths at 30 (VB1), 60 (VB2) and 90 (VB3) m. The experimental results were obtained in cooling and heating seasons of 2006–2007. ANNs have been used in varied applications and they have been shown to be particularly useful in system modeling and system identification. In this study, the back-propagation learning algorithm with three different variants, namely Levenberg–Marguardt (LM), Pola–Ribiere conjugate gradient (CGP), and scaled conjugate gradient (SCG), and tangent sigmoid transfer function were used in the network so that the best approach could be found. The most suitable algorithm and neuron number in the hidden layer were found as LM with 8 neurons for both cooling and heating modes.     

Detection of femoral artery occlusion from spectral density of Doppler signals using the artificial neural network

This research is concentrated on the diagnosis of occlusion disease through the analysis of femoral artery Doppler signals with the help of Artificial Neural Network (ANN). Doppler femoral artery signals belong to occlusion patient and healthy subjects were recorded. Afterwards, power spectral densities (PSD) of these signals were obtained using Welch method and Autoregressive (AR) modeling. Multilayer feed forward ANN trained with a Levenberg Marquart (LM) backpropagation algorithm was implemented to these PSD. The designed classification structure has about 98% sensitivity, 97–100% specifity and correct classification is calculated to be 98–99% (for AR modeling and Welch method respectively). The end results are classified as healthy and diseased. Testing results were found to be compliant with the expected results that are derived from the physician's direct diagnosis. The end benefit would be to assist the physician to make the final decision without hesitation.     

Pruning artificial neural networks: An example using land cover classification of multi-sensor images

The use of three techniques for pruning artificial neural networks (magnitude-based pruning, optimum brain damage and optimal brain surgeon) is investigated, using microwave SAR and optical SPOT data to classify land cover in a test area located in eastern England. Results show that it is possible to reduce network size significantly without compromising overall classification accuracy; indeed, accuracy may rise as the number of links decreases. However, individual class accuracies and the spatial distribution of the pixels forming the individual classes may change significantly. If the network is pruned too severely some classes may be eliminated altogether. In terms of maintaining overall classification accuracy the optimal brain surgeon algorithm gave the best results, and magnitude-based pruning also gave good results despite its simplicity. The optimum brain damage algorithm performed least well of the three methods tested.     

Mitigation of the ground reflection effect in real-time locating systems based on wireless sensor networks by using artificial neural networks

Wireless sensor networks (WSNs) have become much more relevant in recent years, mainly because they can be used in a wide diversity of applications. Real-time locating systems (RTLSs) are one of the most promising applications based on WSNs and represent a currently growing market. Specifically, WSNs are an ideal alternative to develop RTLSs aimed at indoor environments where existing global navigation satellite systems, such as the global positioning system, do not work correctly due to the blockage of the satellite signals. However, accuracy in indoor RTLSs is still a problem requiring novel solutions. One of the main challenges is to deal with the problems that arise from the effects of the propagation of radiofrequency waves, such as attenuation, diffraction, reflection and scattering. These effects can lead to other undesired problems, such as multipath. When the ground is responsible for wave reflections, multipath can be modeled as the ground reflection effect. This paper presents an innovative mathematical model for improving the accuracy of RTLSs, focusing on the mitigation of the ground reflection effect by using multilayer perceptron artificial neural networks.     

基于人工神经网络的含硫原油VGO饱和份含量预测研究

利用多种含硫原油的实验分析数据,采用动最自适应学习率的BP人工神经网络理论,建立了含硫原油VGO的平均沸点、密度、分子量、折光率等4个基础物性与其饱和分含量关系的预测模型,通过该人工神经网络模型的训练,获得了较高的训练精度,模型计算值与实验值相比,VGO饱和份含量的平均相对误差为1.59%;用该模型检验未参加训练的6种油样,预测结果的平均相对误差为5.79%,表明此方法拟合精度较高、预测能力较强,可用于含硫原油基础物性的初步预测.     

Geographical classification of Spanish bottled mineral waters by means of iterative models based on linear discriminant analysis and artificial neural networks

The composition of Spanish natural mineral waters has been determined by means of inductively coupled plasma-mass spectrometry, inductively coupled plasma-atomic emission spectrometry, ionic chromatography and other routine techniques. Methods were applied to samples of bottled water from springs situated in five different mountain systems such as Cordillera Costero-Catalana, Macizo Galaico, Sistemas Béticos, Sistema Central and Sistema Ibérico. Pattern recognition techniques have been applied to differentiate the origin of samples. Data were initially studied by using nonparametric multiple comparison techniques and principal component analysis to highlight data trends. Classification models based on linear discriminant analysis and multilayer perceptron artificial neural networks have been built and validated by means of a stratified jackknifing methodology. An iterative approach has been used to build an artificial neural network model based on the variables selected by linear discriminant analysis. The prediction ability of the constructed model was 94 %.

     

Adaptive classification of temporal signals in fixed-weight recurrent neural networks: an existence proof.

Recurrent neural networks with fixed weights have been shown in practice to successfully classify adaptively signals that vary as a function of time in the presence of additive noise and parametric perturbations. We address the question: Can this ability be explained theoretically? We provide a mathematical proof that these networks have this ability even when parametric perturbations enter the signals nonlinearly. The restrictions that we impose on the signals to be classified are that they satisfy an assumption of nondegeneracy and that noise amplitude is sufficiently small. Further, we demonstrate that the recurrent neural networks may not only classify uncertain signals adaptively but also can recover the values of uncertain parameters of the signals, up to their equivalence classes.     

Retraction Note to: Prediction water absorption resistance of lightweight geopolymers by artificial neural networks

Neural Computing and Applications - The Editor-in-Chief has retracted this article [1] because it significantly overlaps with a number of articles including.     

Generation of optimal artificial neural networks using a pattern search algorithm: application to approximation of chemical systems

A pattern search optimization method is applied to the generation of optimal artificial neural networks (ANNs). Optimization is performed using a mixed variable extension to the generalized pattern search method. This method offers the advantage that categorical variables, such as neural transfer functions and nodal connectivities, can be used as parameters in optimization. When used together with a surrogate, the resulting algorithm is highly efficient for expensive objective functions. Results demonstrate the effectiveness of this method in optimizing an ANN for the number of neurons, the type of transfer function, and the connectivity among neurons. The optimization method is applied to a chemistry approximation of practical relevance. In this application, temperature and a chemical source term are approximated as functions of two independent parameters using optimal ANNs. Comparison of the performance of optimal ANNs with conventional tabulation methods demonstrates equivalent accuracy by considerable savings in memory storage. The architecture of the optimal ANN for the approximation of the chemical source term consists of a fully connected feedforward network having four nonlinear hidden layers and 117 synaptic weights. An equivalent representation of the chemical source term using tabulation techniques would require a 500 x 500 grid point discretization of the parameter space.