Reviews of our study have shown that the complexity of a vague and inaccurate functional relationship can be best ameliorated by fuzzyregression analysis in multivariate environment. However existing algorithm is limited to mainly fuzzy linear functions though literatures have suggested that the concept could be extended to non-linear and intrinsically linear function . This paper is the first attempt of such efforts. In developing the algorithm we have followed the suggestions proposed by related literatures but specifically tailored to the nature of the logistic function which we were focusing at. The algorithm is presented in Fig. 1 below, followed by its algebraic form.
288 www.hrmars.com/journals be trained, this training process is the procedure to obtain the weights of each connection and the neurons threshold value. There are many training algorithms were developed, including the back-propagation (BP) algorithm, the Levenberg Marquardt (LM) and so on. The aim of all these algorithms is to achieve the minimal value of network error. The ANN trained time series are capable to model arbitrarily linear and nonlinear functions. ANN models properties make them utilizing in the various different fields such as time series forecasting and specially it has been used in electricity price forecasting. In most cases an ANN is an adaptive system that changes its structure based on external or internal information that flows through the network during the learning phase. Modern neural networks are non-linear statistical data modeling tools. Basic computational element of model neuron is a node or unit. Its input comes from other units, or from an external source. These inputs together are considered as a vector. Each signal is multiplied to its weight. The sum of these weighted inputs is showed the net input to unit.
In fact, many manufacturing processes tend to be very complex in behavior and have inherent system fuzziness; such as, fluctuation of process pressure and temperature due to environmental effects. This research, therefore, aims at optimizing multiple responses in the manufacturing application on the Taguchi methodusingfuzzyregression analysis.
Cloacal temperature (CT) of broiler chickens is an important parameter to classify its comfort status; therefore its prediction can be used as decision support to turn on acclimatization systems. The aim of this research was to develop and validate a systemusing the fuzzy set theory for CT prediction of broiler chickens. The fuzzysystem was developed based on three input variables: air temperature (T), relative humidity (RH) and air velocity (V). The output variable was the CT. The fuzzy inference system was performed via Mamdani’s method which consisted in 48 rules. The defuzzification was done using center of gravity method. The fuzzysystem was developed using MAPLE ® 8. Experimental results, used for validation, showed
Prediction of stock market returns is an important issue in finance. However, information regarding a stock is normally incomplete, uncertain and vague, making it a challenge to predict the future economic performance. Accurate predictions of stock markets are important for many reasons. Chief among these is the need for investors to hedge against potential market risks and the opportunities for speculators and arbitrators to make profits by trading indexes. Reliability of traditional analysis methods strongly relying on experience is somewhat being doubted due to the complexity of the mode of stock exchange and the correlated information . Techniques such as Regression models and ARIMA models  are also used for stock price forecasting. These models and methods have been used extensively in past. However they failed to give accurate results for some series because of their linear structures and some other limitations. Artificial neural networks (ANN) have been used in stock market prediction during the last decade. Recent research in the area of neural networks has shown that neural networks possess the properties required for relevant applications such as non-linear and smooth interpolations, ability to learn complex non linear mappings and self-adaptation for different statistical distributions. ANN is applied to Tokyo Stock exchange to predict buying and selling signals [2-3]. In
The first attempt of using expert systems for bankruptcy prediction was the one of Messier and Hansen . The objective of the proposed 'data-driven' method was to take firms of known classes (bankrupt/non-bankrupt) described by a fixed set of attributes (financial ratios), and then to generate a production systemusing attributes which correctly classify all the firms of the sample. The rules at each stage (i.e. the variable and the cut-off score) were defined by using measures of entropy and selecting the minimum entropy rule. A decision tree was derived from the production system rules. Messier and Hansen's study was based on a sample of 23 firms (8 bankrupt and 15 non-bankrupt). The classification accuracy of the production system was encouraging in this small case.
Traﬃc volume is a fundamental variable in several transportation engineering applications. For instance, in transportation planning, the annual average daily traﬃc (AADT) is a primary element that has to be estimated for the year of horizon of the analysis. The huge amounts of money to be invested in designed transportation systems are strongly associated with the traﬃc volumes expected in the system, which means that it is important that the AADT should be accurately predicted. In this paper, a modiﬁed version of a pattern recognition technique known as support vector machine for regression (SVR) to forecast AADT is presented. The proposed method- ology computes the SVR prediction parameters based on the distribution of the training data. Therefore, the proposed method is called SVR with data-dependent parameters (SVR-DP). Using 20 years of AADT for both rural and urban roads in 25 counties in the state of Tennessee, the performance of the SVR-DP was compared with those of Holt exponential smoothing (Holt-ES) and of ordinary least- square linear regression (OLS-regression). SVR-DP performed better than both methods; although the Holt-ES also presented good results.
Abstract – Objective: To investigate different fuzzy arithmetical operations to support in the diagnostic of epileptic events and non epileptic events. Method: A neuro-fuzzysystem was developed using the NEFCLASS (NEuro Fuzzy CLASSIfication) architecture and an artificial neural network with backpropagation learning algorithm (ANNB). Results: The study was composed by 244 patients with a bigger frequency of the feminine sex. The number of right decisions at the test phase, obtained by the NEFCLASS and ANNB was 83.60% and 90.16%, respectively. The best sensibility result was attained by NEFCLASS (84.90%); the best specificity result were attained by ANNB with 95.65%. Conclusion: The proposed neuro-fuzzysystem combined the artificial neural network capabilities in the pattern classifications together with the fuzzy logic qualitative approach, leading to a bigger rate of system success.
In contrast, some researchers have used software computing methods developing empirical predictive models for kappa number using different data-driven approaches. An ANN-based strategy for detection of feedstock variations in a continuous pulp digester were studied by Dufour et al. (2005). Wood chip moisture content and densities and alkali and sulfidity in the white liquor were modeled in a pilot plant. Ahvenlampi and Kortela (2005) developed a kappa number prediction model and fault diagnostics of continuous digesters using clustering techniques.The results showed the usability of the combined hybrid system in the monitoring of the process and the kappa number prediction. Halmevaara (2009) developed a novel methodusing multivariate regression to capture the dependencies among the system parameters and quality measures for large industries, presenting results of regression adjustments as an interactive case study simulation of a double vessel softwood pulp continuous digester. Araneda et al. (2009) adapted the Purdue model to the physical characteristics of a Kamyr digester. This model was able to represent satisfactorily both dynamic and steady states of the digester operation, improving information from previous models. Predicted data obtained from this model were compared to measured ones from mills, such as blow-line kappa number, yield, free liquor temperature profile, and pulp production rate. Saavedra (2011) selected 29 cooking variables from his experience with a continuous digester, and used a MLR and ANN for predictive models, concluding that the ANN presented better results. Galicia et. al (2012) applied soft sensors using secondary measurements based on multivariate regression techniques. They developed a software sensor in order to reduce the number of regressor variables and also to provide superior prediction performance of kappa number applied in both simulated and industrial continuous Kamyr digester case studies. Kraft pulping has been a widely studied subject, especially concerning softwood pulp. Nevertheless, there are only a few references to kappa number prediction techniques concerning statistical and artificial neural network models from industrial hardwood pulping data. In this sense, this work brings an important contribution to the studies involving hardwood processing.
fundamental theories of basic sciences and give good understanding about the mechanics of the process. However, the prediction accuracy of mathematical models is not very encouraging due to the assumptions or simplifications used while building these models. Statistical regression models proposed by Hafez , Hunter , Mogahzy , and Smith and Waters  are very simple to understand and the beta coefficient analysis gives an indication of relative importance of various inputs on the yarn strength. However, foretelling the type of relationship (liner or non-linear) is essential for developing a regression model. The advent of artificial intelligence has provided a new impetus in the research on modelling of yarn properties. Cheng and Adams , Ramesh, Rajamanickam and Jayaraman, , Zhu and Ethridge [12, 13], Guha, Chattopadhyay and Jayadeva  and Majumdar and Majumdar  have successfully used the artificial neural network (ANN) and neural-fuzzy methods to predict various properties of spun yarns. The prediction accuracy of ANN has been acclaimed by most of these researchers. However, ANN modelling has also received criticisms galore for acting like a ‘black box’ without revealing much about the mechanics of the process.
In this paper, a numerical method for integration of fuzzy functions is considered. Fuzzy Newton-Cotes formula, such as fuzzy trapezoidal method and fuzzy Simpson method are calculated by integration of fuzzy functions on two and three equally space points. Also the composite fuzzy trapezoidal and composite fuzzy Simpson method are proposed for n equally space points. The proposed method are illustrated by numerical examples.
The compounds included in each set are speciied in Table 1. The six simulant/polymer/migrant descriptors were used as inputs for development of the ANFIS model. The model building involves two stages: structure identiication and parameter identiication. The former is related to inding a suitable number of rules and a proper partition of the feature space. The latter is concerned with the adjustment of system parameters, such as MF (membership function) parameters, linear coeficients, and so on. It is concluded that by increasing the number of MFs per input, the number of rules increases accordingly. For the irst stage of ANFIS modeling, grid partitioning should be used for partitioning the features. The number and type of membership functions should be optimized by using RMSE as a criterion for the test set. All ANFIS models were produced using MATLAB 7.0 Fuzzy Logic Toolbox (MATLAB, Mathworks Inc. software, Natick, USA, 2008).
Lets suppose that, in a teaching environment, a student finds at his disposal a great number of learning objects, such as practices or exercises. The student has at his disposal many more objects than he is able to use, and has no idea where he should begin, so bearing in mind the learning objects are classified by categories, he decides to begin with the basic level. The student browses through all these educational objects for their topics and remembers a friend told him how much he had liked those learning objects related with an specific topic. The student decides to start with those objects, and once he has finished with them he calls his friend so he can recommend him more since the ones he has already gone trough did match what he was looking for.
using BP/TZVP and the COSMOtherm continuum solvation model. The gas phase deprotonation standard free energy is computed without vibrational corrections and the pKa values are computed using Eq. (2), where A and B are found by a linear fit to experimental pKa values for a training set of 43 amines. Eckert & Klamt (2005) observed that the method systematically underestimates the pKa of secondary and tertiary aliphatic amines by ca 1 and 2 pH units, respectively, so an additional empirical correction is added for these two molecule types. Using this approach the pKa values of 58 drug-like molecules containing one or more ionizable N atoms can be reproduced with a root mean square deviation (RMSD) of 0.7 pH units.
However existing researchers concentrate on source, destination and neighbor recovery by activating all sleeping sensors to find out the missing object. If this case fails, which lead to flooding recovery (i.e.) wakes up all the nodes in the network and put the network in high energy consumption. Still Some of the factors that impact the energy consumption in object tracking they are number of moving objects, reporting frequency, data precision, sampling frequency, object moving speed, location models.. In order to overcome this situation we propose a method known as SMCM to localize the missing object, when the object is not found by the sensor nodes during object tracking and at last we compare the simulated results of SMCM with the multilatertion and centroid methods, to visualize the result. We used here two metrics for performance evaluation such as network energy consumption, and localization error. Differ from other researches; our aim of this paper is to improve the energy efficiency and to minimize the localization error by using the proposed technique to extend the lifetime of the network.
A kernel is a function that enables the support vector machine to linearly separate the data in a higher-dimensional space . Using a kernel function is similar to adding a trivial feature to the already existing set of features only except no feature is added or removed. Kernel function K(x, y) can be expressed as a dot product in a high dimensional space. If the arguments to the kernel are in a measurable space X,
st udy w it h m ult iple pr ogr am s m ay r ev eal differences in how pr ogr am s v iew lear ning env ir onm ent s as a m eans t o incr ease lear ning. I t w ould also be int er est ing t o invest igat e how inst r uctors could im plem ent changes to cour ses based on dat a fr om each adm inist r at ion of t he CLS. Adding open ended quest ions t o t he CLS for st udent s and inst r uct ors m ay shed addit ional insight on learning. St udent s could be ask ed t o descr ibe how t heir com for t abilit y w it h classm at es, inst r uct or , and course cont ent changed t hroughout t he sem est er and inst r uctor s could be asked t o descr ibe how t hey used t he dat a for m t he CLS to infor m t heir t eaching.
Following previous research for this river, two abiotic inputs (daily mean water tem- perature, T and daily mean flow rate, Q) will be used to predict daily minimum DO. An advantage of using these factors is that they are routinely collected by the City of Calgary, and thus, a large dataset is available. Also, their use in previous studies has shown that they are good predictors of daily DO concentration in this river basin (He
solutions, which is the usual form to obtain a solution without explicitly considering the phenomenon called wavebreaking, in which past a certain point the peak of the wave moves faster, creating a multiple valued solution. Figure (3.3) shows the loss function for the training of the problem using the DGM architecture. It falls sharply right at the first iterations, and stops improving after around 200 iterations. For the viscid form we will consider the initial-value problem
An ANFIS model characterizing a flexible plate structure has been presented and discussed. The best feature of ANFIS is that it pre-processes all the data into several membership functions before mapping the data into an adaptive neuro structure. This pre-processing feature allows ANFIS to converge faster and better .A dynamic model based on experimental study is developed at initial stage of the work. The ANFIS model has been evaluated and the ANFIS output is plotted against the actual output. ANFIS model has been shown to exhibit like the true system, with small errors developed between actual and predicted output. This model can be used to in future work to design vibration suppression mechanism on the flexible structure.