Stock market prediction using artificial neural networks with optimal feature transformation

This paper compares a feature transformation method using a genetic algorithm (GA) with two conventional methods for artificial neural networks (ANNs). In this study, the GA is incorporated to improve the learning and generalizability of ANNs for stock market prediction. Daily predictions are conducted and prediction accuracy is measured. In this study, three feature transformation methods for ANNs are compared. Comparison of the results achieved by a feature transformation method using the GA to the other two feature transformation methods shows that the performance of the proposed model is better. Experimental results show that the proposed approach reduces the dimensionality of the feature space and decreases irrelevant factors for stock market prediction.     

Cooperative binary-real coded genetic algorithms for generating and adapting artificial neural networks

This paper describes a genetic system for designing and training feed-forward artificial neural networks to solve any problem presented as a set of training patterns. This system, called GANN, employs two interconnected genetic algorithms that work parallelly to design and train the better neural network that solves the problem. Designing neural architectures is performed by a genetic algorithm that uses a new indirect binary codification of the neural connections based on an algebraic structure defined in the set of all possible architectures that could solve the problem. A crossover operation, known as Hamming crossover, has been designed to obtain better performance when working with this type of codification. Training neural networks is also accomplished by genetic algorithms but, this time, real number codification is employed. To do so, morphological crossover operation has been developed inspired on the mathematical morphology theory. Experimental results are reported from the application of GANN to the breast cancer diagnosis within a complete computer-aided diagnosis system.     

Hybrid control of the three phase induction machine using artificial neural networks and fuzzy logic

Nowadays, the microcomputer performs calculations at an incredibly high rate of billions of instructions per second. That represents an exponential increase in the processing speed since the early days of the computer development, eventhough such growth did not show complex reasoning that even the simple biological organisms can make. The artificial intelligence techniques as an attempt to work about those limitations, are a promising alternative.Each intelligent technique has its particular strengths and weaknesses and cannot be universally implemented to any problem. Mixed together, these techniques can improve the solutions quality and allow application to various tasks. It is the reason why the AI is used increasingly in order to solve complex problems in engineering. Where, it is still necessary to make progress in the controller tuning.The idea proposed in this paper is simple and original. It is the result of a study that compared the performances of two controls based on the artificial intelligence techniques: the artificial neural networks and the fuzzy logic. The control proposed in this paper combines in a different manner these two techniques in the form of a hybrid control. The aim is to benefit from performances of each of these techniques, by using them in the same control block at the most suitable place.The performances of the this proposed hybrid control; applied to the three-phase induction motor supplied by voltage source inverter; are investigated and compared to those obtained from the controls based on artificial neural networks; fuzzy logic and conventional techniques. The results of simulation show the feasibility and the good performances achieved by the proposed control.     

Multi-objective optimization design of a complex building based on an artificial neural network and performance evaluation of algorithms

While optimization studies focusing on real-world buildings are somewhat limited, many building optimization studies to date have used simple hypothetical buildings for the following three reasons: (1) the shape and form of real buildings are complex and difficult to mathematically describe; (2) computer models built based on real buildings are computationally expensive, which makes the optimization process time-consuming and impractical and (3) although algorithm performance is crucial for achieving effective building performance optimization (BPO), there is a lack of agreement regarding the proper selection of optimization algorithms and algorithm control parameters. This study applied BPO to the design of a newly built complex building. A number of design variables, including the shape of the building’s eaves, were optimized to improve building energy efficiency and indoor thermal comfort. Instead of using a detailed simulation model, a surrogate model developed by an artificial neural network (ANN) was used to reduce the computing time. In this study, the performance of four multi-objective algorithms was evaluated by using the proposed performance evaluation criteria to select the best algorithm and parameter values for population size and number of generations. The performance evaluation results of the algorithms implied that NSGA-II (with a population size and number of generations of 40 and 45, respectively) performed the best in the case study. The final optimal solution significantly improves building performance, demonstrating the success of the BPO technique in solving complex building design problems. In addition, the findings on the performance evaluation of the algorithms provide guidance for users regarding the selection of suitable algorithms and parameter settings based on the most important performance criteria.     

Least-cost design of singly and doubly reinforced concrete beam using genetic algorithm optimized artificial neural network based on Levenberg–Marquardt and quasi-Newton backpropagation learning techniques

In this work, least-cost design of singly and doubly reinforced beams with uniformly distributed and concentrated load was done by incorporating actual self-weight of beam, parabolic stress block, moment–equilibrium and serviceability constraint besides other constraints. Also, this design expertise was incorporated into a genetically optimized artificial neural network based on steepest descent, Levenberg–Marquardt, and quasi-Newton backpropagation learning techniques. The initial solution for the optimization procedure was obtained using limit state design as per IS: 456-2000.     

Track geometry prediction for Laser Metal Deposition based on on-line artificial vision and deep neural networks

Laser Metal Deposition (LMD) is an additive manufacturing technology that attracts great interest from the industry, thanks to its potential to realize parts with complex geometries in one piece, and to repair damaged ones, while maintaining good mechanical properties. Nevertheless, the complexity of this process has limited its widespread adoption, since different part geometries, strategies and boundary conditions can yield very different results in terms of external shapes and inner flaws. Moreover, monitoring part quality during the process execution is very challenging, as direct measurements of both structural and geometrical properties are mostly impracticable. This work proposes an on-line monitoring and prediction approach for LMD that exploits coaxial melt pool images, together with process input data, to estimate the size of a track deposited by LMD. In particular, a novel deep learning architecture combines the output of a convolutional neural network (that takes melt pool images as inputs) with scalar variables (process and trajectory data). Various network architectures are evaluated, suggesting to use at least three convolutional layers. Furthermore, results imply a certain degree of invariance to the number and size of dense layers. The effectiveness of the proposed method is demonstrated basing on experiments performed on single tracks deposited by LMD using powders of Inconel 718, a relevant material for the aerospace and automotive sectors.     

Reverse analysis via efficient artificial neural networks based on simulated Berkovich indentation considering effects of friction

Instrumented indentation test has become a popular method for characterization of materials of small volume such as those constitute the micro-electro-mechanical devices, micro-electronic packages and thin film. Berkovich indenter is one of the most popular indenter tips employed in the tests. The present study involves the finite element simulation of indentation by Berkovich-family of indenters to establish the load-displacement relations for elasto-plastic materials obeying power law. Effects of friction at the contact surfaces, which have been ignored by most of the researchers are considered in the analyses. Extensive 3-dimensional finite element analyses covering a wide practical range of materials have been carried out and the results adopted for material characterization via artificial neural network model based on an efficient reverse analysis algorithm. Direct mapping of the characteristics of the indentation curves to the material properties are performed and the characteristics of the network model deliberated. The tuned network can then be adopted to predict the mechanical properties of a new set of materials of small volume in micro-electro-mechanical components.     

The experimental investigation of the effects of uncoated,PVD- and CVD-coated cemented carbide inserts and cutting parameters on surface roughness in CNC turning and its prediction using artificial neural networks

In this study the machining of AISI 1030 steel (i.e. orthogonal cutting) uncoated, PVD- and CVD-coated cemented carbide insert with different feed rates of 0.25, 0.30, 0.35, 0.40 and 0.45 mm/rev with the cutting speeds of 100, 200 and 300 m/min by keeping depth of cuts constant (i.e. 2 mm), without using cooling liquids has been accomplished. The surface roughness effects of coating method, coating material, cutting speed and feed rate on the workpiece have been investigated. Among the cutting tools—with 200 mm/min cutting speed and 0.25 mm/rev feed rate—the TiN coated with PVD method has provided 2.16 μm, TiAlN coated with PVD method has provided 2.3 μm, AlTiN coated with PVD method has provided 2.46 μm surface roughness values, respectively. While the uncoated cutting tool with the cutting speed of 100 m/min and 0.25 mm/rev feed rate has yielded the surface roughness value of 2.45 μm. Afterwards, these experimental studies were executed on artificial neural networks (ANN). The training and test data of the ANNs have been prepared using experimental patterns for the surface roughness. In the input layer of the ANNs, the coating tools, feed rate (f) and cutting speed (V) values are used while at the output layer the surface roughness values are used. They are used to train and test multilayered, hierarchically connected and directed networks with varying numbers of the hidden layers using back-propagation scaled conjugate gradient (SCG) and Levenberg–Marquardt (LM) algorithms with the logistic sigmoid transfer function. The experimental values and ANN predictions are compared by statistical error analyzing methods. It is shown that the SCG model with nine neurons in the hidden layer has produced absolute fraction of variance (R²) values about 0.99985 for the training data, and 0.99983 for the test data; root mean square error (RMSE) values are smaller than 0.00265; and mean error percentage (MEP) are about 1.13458 and 1.88698 for the training and test data, respectively. Therefore, the surface roughness value has been determined by the ANN with an acceptable accuracy.     

Generalization performance of support vector machines and neural networks in runoff modeling

Effective one-day lead runoff prediction is one of the significant aspects of successful water resources management in arid region. For instance, reservoir and hydropower systems call for real-time or on-line site-specific forecasting of the runoff. In this research, we present a new data-driven model called support vector machines (SVMs) based on structural risk minimization principle, which minimizes a bound on a generalized risk (error), as opposed to the empirical risk minimization principle exploited by conventional regression techniques (e.g. ANNs). Thus, this stat-of-the-art methodology for prediction combines excellent generalization property and sparse representation that lead SVMs to be a very promising forecasting method. Further, SVM makes use of a convex quadratic optimization problem; hence, the solution is always unique and globally optimal. To demonstrate the aforementioned forecasting capability of SVM, one-day lead stream flow of Bakhtiyari River in Iran was predicted using the local climate and rainfall data. Moreover, the results were compared with those of ANN and ANN integrated with genetic algorithms (ANN-GA) models. The improvements in root mean squared error (RMSE) and squared correlation coefficient (R²) by SVM over both ANN models indicate that the prediction accuracy of SVM is at least as good as that of those models, yet in some cases actually better, as well as forecasting of high-value discharges.     

Predicting body fat percentage based on gender,age and BMI by using artificial neural networks

In the human body, the relation between fat and fat-free mass (muscles, bones etc.) is necessary for the diagnosis of obesity and prediction of its comorbidities. Numerous formulas, such as Deurenberg et al., Gallagher et al., Jackson and Pollock, Jackson et al. etc., are available to predict body fat percentage (BF%) from gender (GEN), age (AGE) and body mass index (BMI). These formulas are all fairly similar and widely applicable, since they provide an easy, low-cost and non-invasive prediction of BF%. This paper presents a program solution for predicting BF% based on artificial neural network (ANN). ANN training, validation and testing are done by randomly divided dataset that includes 2755 subjects: 1332 women (GEN = 0) and 1423 men (GEN = 1), with AGE from 18 to 88 y and BMI from 16.60 to 64.60 kg/m². BF% was estimated by using Tanita bioelectrical impedance measurements (Tanita Corporation, Tokyo, Japan). ANN inputs are: GEN, AGE and BMI, and output is BF%. The predictive accuracy of our solution is 80.43%. The main goal of this paper is to promote a new approach to predicting BF% that has same complexity and costs but higher predictive accuracy than above-mentioned formulas.     

A comparison of gradient ascent, gradient descent and genetic-algorithm-based artificial neural networks for the binary classification problem

Abstract: We compare log maximum likelihood gradient ascent, root-mean-square error minimizing gradient descent and genetic-algorithm-based artificial neural network procedures for a binary classification problem. We use simulated data and real-world data sets, and four different performance metrics of correct classification, sensitivity, specificity and reliability for our comparisons. Our experiments indicate that a genetic-algorithm-based artificial neural network that maximizes the total number of correct classifications generally fares well for the binary classification problem. However, if the training data set contains inconsistent decisions or noise then the log maximum likelihood maximizing gradient ascent may be the best classification approach to use. The root-mean-square minimizing gradient descent approach appears to overfit training data and has the lowest reliability among the approaches considered for our research. At the end of the paper, we provide a few guidelines, including computational complexity, for selection of an appropriate technique for a given binary classification problem.     

Evidences of cognitive effects over auditory steady-state responses by means of artificial neural networks and its use in brain–computer interfaces

The auditory steady-state response is an EEG potential elicited by the repetitive presentation of auditory stimuli. Researchers have found contradictory results about the influence of cognitive tasks, such as the selective attention, over this potential. It has been proved that selective attention is able to modulate cortex originated steady-state responses, such as the visual. This fact has been widely used to develop brain–computer interfaces. However for complete locked-in patients, such as those in an advanced state of Amyotrophic lateral sclerosis, visual stimuli are not longer suitable, hence the need of another type of stimulus, generally auditory, for both stimulation and feedback. In this paper we present a study based on artificial neural networks that evidences the effects of selective attention over auditory steady-state responses and the use in brain–computer interfaces is discussed.