<title>Feature space trajectory (FST) classifier neural network</title>

There is a lack of literature about the classification performance improvement effect of hyperparameter tuning to predict health expenditure per capita (HE). In this study, the effect of hyperparameter tuning on classification performances of random forest (RF) and neural network (NN) classification tasks is compared for grouping member of World Bank (WB) countries in terms of HE. Data gathered from 188 member countries of WB for the year 2019. GDP per capita, mortality, life expectancy at birth and population aged 65 years and over are used as predictors. Number of trees and neurons in hidden layer are changed from 5 to 100 for RF and NN by changing k-fold parameter from 2 to 20. The dependent HE variable is transformed into binary categories, and the categories are well balanced (%50–%50). Classification performances of learning techniques are good (AUC > 0.95). RF (AUC = 0.9609) is superior to NN (AUC = 0.9596) in terms of average AUC values generated by hyperparameter tuning.

Historical database training method for neural networks

We advance active computer vision algorithms for flexible manufacturing systems that classify objects and estimate their pose from intensity images. Our algorithms automatically reposition the sensor if the class or pose of an object is ambiguous in a given image and incorporate data from multiple object views in determining the final object classification. A feature space trajectory (FST) in a global eigenfeature space is used to represent 3-D distorted views of an object. Bayesian methods are used to derive the class hypothesis, pose estimate, confidence measures, and the sensor position that best resolves ambiguity. FSTs constructed using images rendered from solid models of objects are used to process real image data.

We advance new active computer vision algorithms that classify objects and estimate their pose from intensity images. Our algorithms automatically reposition the sensor if the class or pose of an object is ambiguous in a given image and incorporate data from multiple object views in determining the final object classification. A feature space trajectory (FST) in a global eigenfeature space is used to represent 3-D distorted views of an object. Assuming that an observed feature vector consists of Gaussian noise added to a point on the FST, we derive a probability density function (PDF) for the observation conditioned on the class and pose of the object. Bayesian estimation and hypothesis testing theory are then used to derive approximations to the maximum a posteriori probability pose estimate and the minimum probability of error classifier. New confidence measures for the class and pose estimates, derived using Bayes theory, determine when additional observations are required as well as where the sensor should be positioned to provide the most useful information.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Although flexible neural networks (FNNs) have been used more successfully than classical neural networks (CNNs), nothing is rigorously known about their properties. In fact, they are not even well known to the systems and control community. In this paper, theoretical evidence is given for their superiority over CNNs. Following an overview of flexible bipolar sigmoid functions (FBSFs), several fundamental properties of feedforward and recurrent FNNs are established. For the feedforward case, it is proven that similar to CNNs, FNNs with as few as a single hidden layer (SHL) are universal approximators. It is also proven that unlike irreducible SHL CBSNNs, irreducible SHL FBSNNs are nonuniquely determined by their input-output (I-0) maps, up to a finite group of symmetries. Then, recurrent FNNs are introduced. It is observed that they can be interpreted as a generalization of the conventional state-space framework. For the recurrent case, it is substantiated that similar to CBSNNs, FBSNNs are universal approximators. Necessary and sufficient conditions for the controllability and observability of a generic class of them are established. For a subclass of this class, it is proven that unlike CBSNNs, FBSNNs are nonuniquely determined by their I-O maps, up to a finite group of symmetries, and that every system inside this subclass is minimal. Finally, a new class of FNNs, namely, flexible bipolar radial basis neural networks (FBRBNNs) is introduced. It is proven that as in the case of classical radial basis neural networks (CRBNNs), feedforward SHL FBRBNNs are universal approximators.

Deep learning algorithms can be used to design systems that report data on patients and deliver warnings to medical applications or electronic health information if there are changes in the patient's health. These systems could be created using deep Learning. This helps verify that patients get the proper effective care at the proper time for each specific patient. A healthcare decision support system was presented using the Internet of Things and deep learning methods. In the proposed system, we examined the capability of integrating deep learning technology into automatic diagnosis and IoT capabilities for faster message exchange over the Internet. We have selected the suitable Neural Network structure (number of best-hidden layers and activation function classes) to construct the e-health system. In addition, the e-health system relied on data from doctors to understand the Neural Network. In the validation method, the total evaluation of the proposed healthcare system for diagnostics provides dependability under various patient conditions. Based on evaluation and simulation findings, a dual hidden layer of feed-forward NN and its neurons store the tanh function more effectively than other NN. To overcome challenges, this study will integrate artificial intelligence with IoT. This study aims to determine the NN's optimal layer counts and activation function variations.

With a focus on classification problem, in this paper, we present an integrated approach to improve the performance of classification using adaptive resonance theory (ART) neural network and logistic regression classifiers. In our approach, the neural network classifier is trained first and then regression analysis is applied to each individual class. In testing phase, the data is applied to the regression classifier and, if any deviation exists, the neural network classifier is retrained. The study reveals that effective data mining can be achieved by combining the power of neural networks with the rigor of more traditional statistical tools.

Researchers in the fields of neural networks, statistics, machine learning, and artificial intelligence have followed three basic approaches to developing new pattern classifiers. Probability Density Function (PDF) classifiers include Gaussian and Gaussian Mixture classifiers which estimate distributions or densities of input features separately for each class. Posterior probability classifiers include multilayer perceptron neural networks with sigmoid nonlinearities and radial basis function networks. These classifiers estimate minimum-error Bayesian a posteriori probabilities (hereafter referred to as posterior probabilities) simultaneously for all classes. Boundary forming classifiers include hard-limiting single-layer perceptrons, hypersphere classifiers, and nearest neighbor classifiers. These classifiers have binary indicator outputs which form decision regions that specify the class of any input pattern. Posterior probability and boundary-forming classifiers are trained using discriminant training. All training data is used simultaneously to estimate Bayesian posterior probabilities or minimize overall classification error rates. PDF classifiers are trained using maximum likelihood approaches which individually model class distributions without regard to overall classification performance. Analytic results are presented which demonstrate that many neural network classifiers can accurately estimate posterior probabilities and that these neural network classifiers can sometimes provide lower error rates than PDF classifiers using the same number of trainable parameters. Experiments also demonstrate how interpretation of network outputs as posterior probabilities makes it possible to estimate the confidence of a classification decision, compensate for differences in class prior probabilities between test and training data, and combine outputs of multiple classifiers over time for speech recognition.

Cognitive Radio (CR) networks has been presented in the last decade as a solution to the problem of increasing congestion in frequency spectrum, through opportunistic spectrum access techniques. One of the main components of Cognitive Radio receivers is the Automatic Modulation Classification (AMC), in which the CR can blindly identify the modulation scheme of a detected signal. AMC has several applications, including military, spectrum surveillance and management, and commercial applications. In this paper we propose an AMC based on two different Neural Networks (NN) classifiers: Feed-Forward with Resilient Back-Propagation NN and Probabilistic NN. NN classifiers take their inputs as a feature vector from a Features extraction phase. Features selected for classification are the statistical features of the received signal's instantaneous amplitude, frequency and phase. Simulations show that both NN classifiers can achieve over 80% correct classification up to Signal-to-Noise Ratio (SNR) of 5 dB, and the probability of correct classification increases up to 99% at a SNR of 15 dB.

A suitable framework for the development of artificial neural networks is important because it decides the level of accuracy, which can be reached for a certain dataset and increases the certainty about the reached classification results. In this paper, we conduct a comparative study for the performance of four frameworks, Keras with TensorFlow, Pytorch, TensorFlow, and Cognitive Toolkit (CNTK), for the elaboration of neural networks. The number of neurons in the hidden layer of the neural networks is varied from 8 to 64 to understand its effect on the performance metrics of the frameworks. A test dataset is synthesized using an analytical model and real measured impedance spectra by an eddy current sensor coil on EUR 2 and TRY 1 coins. The dataset has been extended by using a novel method based on interpolation technique to create datasets with different difficulty levels to replicate the scenario with a good imitation of EUR 2 coins and to investigate the limit of the prediction accuracy. It was observed that the compared frameworks have high accuracy performance for a lower level of difficulty in the dataset. As the difficulty in the dataset is raised, there was a drop in the accuracy of CNTK and Keras with TensorFlow depending upon the number of neurons in the hidden layers. It was observed that CNTK has the overall worst accuracy performance with an increase in the difficulty level of the datasets. Therefore, the major comparison was confined to Pytorch and TensorFlow. It was observed for Pytorch and TensorFlow with 32 and 64 neurons in hidden layers that there is a minor drop in the accuracy with an increase in the difficulty level of the dataset and was above 90% until both the coins were 80% closer to each other in terms of electrical and magnetic properties. However, Pytorch with 32 neurons in the hidden layer has a reduction in model size by 70% and 16.3% and predicts the class, 73.6% and 15.6% faster in comparison to TensorFlow and Pytorch with 64 neurons.

It is a necessity to determine significant food or traditional Chinese medicine (TCM) with low cost, which is more likely to achieve high accurate identification by THz-TDS. In this study, feedforward neural networks based on terahertz spectra are employed to predict the animal origin of gelatins, whose adaption to the mission is examined by parallel models built by random sample partition and initialization. It is found that the generalization performance of feedforward ANNs in original data is not satisfactory although prediction on trained samples can be accurate. A multivariate scattering correction is conducted to enhance prediction accuracy, and 20 additional models verify the effectiveness of such dispose. A special partition of total dataset is conducted based on statistics of parallel models, whose influence on ANN performance is investigated with another 20 models. The performance of the models is unsatisfactory because of notable differences in training and test sets according to principal component analysis. By comparing the distribution of the first two principal components before and after multivariate scattering correction, we found that the reciprocal of the minimum number of line segments required for error-free classification in 2-D feature space can be viewed as an index to describe linear separability of data. The rise of proposed linear separability would have a lower requirement for harsh parameter tuning of ANN models and tolerate random initialization. The difference in principal components of samples between a training set and a data set determines whether partition is acceptable or whether a model would have generality. A rapid way to estimate the performance of an ANN before sufficient tuning on a classification mission is to compare differences between groups and differences within groups. Given that a representative peak missing curve is discussed in this article, an analysis based on gelatin THz spectra may be helpful for studies on some other feature-less species.

Neural network, a nonalgorithmic, nondigital, intensely parallel and distributive information processing system, is being used more and more everyday. The main interest in neural networks is rooted in the recognition that the human brain processes information in a different manner than conventional digital computers. Computers are extremely fast and precise at executing sequences of instructions that have been formulated for them. A human information processing system is composed of neurons switching at speeds about a million times slower than computer gates. Yet, humans are more efficient than computers at such computationally complex tasks as speech and other pattern-recognition problems. Artificial neural systems, or neural networks, are physical cellular systems that can acquire, store, and use experiential knowledge. The knowledge is in the form of stable states or mapping embedded in networks that can be recalled in response to the presentation of cues. In a typical neural data processing procedure, the database is divided into two separate portions called training and test sets. The training set is used to develop the desired network. In this process (depending on the paradigm that is being used), the desired output in the training set is used to help the network learn by adjusting the weights between its neurons or processing elements. Once the network has learned the information in the training set and has"converged," the test set is applied to the network for verification. It is important to note that although the user has the desired output of the test set, it has not been seen by the network. This ensures the integrity and robustness of the trained network. A handful of articles on the use of neural networks in the petroleum industry has appeared in SPE conferences, proceedings, and publications in the past 2 years. These articles can be divided into two categories: those that use neural networks to analyze formation lithology from well logs and those that use neural networks to pick a reservoir model to be used in conventional welltest interpretation studies. These tasks are usually done by log analysts and reservoir engineers, and their automation using a fault-tolerant process may prove valuable. Neural networks can help engineers and researchers by addressing some fundamental petroleum engineering problems as well as specific ones that conventional computing has been unable to solve. Petroleum engineers may benefit from neural networks on occasions when engineering data for design and interpretations are less than adequate. This is an especially common occurrence in the Appalachian basin, where some fields are quite old. Lack of adequate engineering data may also be encountered because of the high cost of coring, well testing, and so on. Neural networks have shown great potential for generating accurate analysis and results from large amounts of historical data that otherwise would seem not to be useful or relevant in the analysis. An example of such a problem was encountered by a gas company for a gas storage field in Ohio. In the absence of appropriate data, which normally would make engineering design and evaluation of the fracturing jobs virtually impossible, a carefully designed neural network was able to predict the performance of fracturing jobs with great accuracy. A linear plot of the actual fracturing job results (data never seen by the network during training) and network predictions resulted in a correlation coefficient of 0.98, where 1.00 is a perfect match. Neural networks have proved to be valuable pattern-recognition tools. They are capable of finding highly complex patterns within large amounts of data. A relevant example is well log interpretation. It is generally accepted that there is more information embedded in well logs than meets the eye. Determination, prediction, or estimation of formation permeability without actual laboratory measurement of the cores or interruption in production for well test data collection has been a fundamental problem for petroleum engineers. From geophysical well log data, it was possible to predict and/or estimate permeability of a highly heterogeneous formation in West Virginia, as shown in Fig 1. Reference Mohaghegh, S., Arefi, R., and Ameri, S.: "Design and Development of an Artificial Neural Network for Prediction of Formation Permeability,"paper SPE 28237 presented at the 1994 SPE Petroleum Computer Conference, July31-Aug. 3, Dallas.

Transfer learning aims to transfer knowledge from the labelled source domain data to build a good classifier for the target domain data which has few labels or none. Existing feature-based transfer learning methods seek to transform the data to a new feature space under which the distribution discrepancy is reduced. However, data in different classes may not be easy to be separated in the new feature representation. Therefore, a modified transfer learning algorithm with Joint Distribution Adaptation (JDA) and Maximum Margin Criterion (MMC) is put forward in this paper, which we call MMC-JDA for short. MMC-JDA is aimed at minimizing the distribution discrepancy between two domains and maximizing the separability of each class at the same time. Comparative experiments on 16 cross-domain public image datasets show that MMC-JDA is effective and performs better than several common transfer learning methods.

The artificial neural network has been proved among the best tools in data mining for classification tasks. The concept of obtaining more accurate classifier with less computational complexity has been gaining importance, because of day by day increase in the data. Several numbers of models have been developed for classification problems. This paper is the depiction of higher order neural networks especially Fibonacci Functional Link Neural Network (FFLNN) for data classification. The coefficients of individual terms in Fibonacci polynomials are smaller than those of individual terms in the classical orthogonal polynomials. Additionally, less number of terms make it a preferable classifier regarding functional expansion. These properties lead this FFLNN to produce more accurate higher order neural network with less computational complexity to tackle the classification problems. Four datasets were collected from KEEL and LIBSVM dataset repositories. Computational results were compared with three benchmarked models including Chebyshev Functional Link Neural Network (CFLNN), Chebyshev Multilayer Perceptron (CMLP) and Multilayer Perceptron Neural Network (MLP). A t-test was applied to check the significance of the proposed classifier based on classification performance. The findings showed that the proposed classifier outperformed all benchmarked models in all evaluation measures.

Context: Classification of erythrocyte morphological changes is usually done by an expert through direct observation from the microscope based on qualitative criteria, leading to subjective diagnosis. Proposals to automate this process usually classified erythrocytes in normal or abnormal, without specifying the type of abnormality that indicates the presence of some disease. We develop a tool for diagnostic support that determines different pathological forms of erythrocytes using characteristics measured from the microscopic image. Method: We detect isolated erythrocytes using a segmentation processes based on color. Then we measure some differentiating features in each cell, including a new measure of central pallor. These features are presented to a neural network that labels the cell according to seven types of abnormality. Results: The resulting system achieves a high success rate (97.3%) compared to binary classifications found in the literature. The measure of central pallor is highly discriminant because it allows a perfect distinction between normocytes and spherocytes, when other morphological characteristics are very similar between them. Conclusions: Our contribution includes the classification of multiple types of erythrocytes and the proposal of a highly discriminating measure of the central pallor. We verified the usefulness of combining pre-processing techniques for extracting features and neural networks for classification in feature space. For future work, it would be desirable to have a greater number of images with statistically significant samples of other types of erythrocytes to verify the goodness of the proposed methodology to classify more cell types. Also, with a greater number of classified samples, different pattern classification techniques could be studied in order to evaluate, compare and select the most appropriate technique.