Effective diagnosis of heart disease through neural networks ensembles

In the last decades, several tools and various methodologies have been proposed by the researchers for developing effective medical decision support systems. Moreover, new methodologies and new tools are continued to develop and represent day by day. Diagnosing of the heart disease is one of the important issue and many researchers investigated to develop intelligent medical decision support systems to improve the ability of the physicians. In this paper, we introduce a methodology which uses SAS base software 9.1.3 for diagnosing of the heart disease. A neural networks ensemble method is in the centre of the proposed system. This ensemble based methods creates new models by combining the posterior probabilities or the predicted values from multiple predecessor models. So, more effective models can be created. We performed experiments with the proposed tool. We obtained 89.01% classification accuracy from the experiments made on the data taken from Cleveland heart disease database. We also obtained 80.95% and 95.91% sensitivity and specificity values, respectively, in heart disease diagnosis.     

Analysing the localisation sites of proteins through neural networks ensembles

Scientists involved in the area of proteomics are currently seeking integrated, customised and validated research solutions to better expedite their work in proteomics analyses and drug discoveries. Some drugs and most of their cell targets are proteins, because proteins dictate biological phenotype. In this context, the automated analysis of protein localisation is more complex than the automated analysis of DNA sequences; nevertheless the benefits to be derived are of same or greater importance. In order to accomplish this target, the right choice of the kind of the methods for these applications, especially when the data set is drastically imbalanced, is very important and crucial. In this paper we investigate the performance of some commonly used classifiers, such as the K nearest neighbours and feed-forward neural networks with and without cross-validation, in a class of imbalanced problems from the bioinformatics domain. Furthermore, we construct ensemble-based schemes using the notion of diversity, and we empirically test their performance on the same problems. The experimental results favour the generation of neural network ensembles as these are able to produce good generalisation ability and significant improvement compared to other single classifier methods.     

Evolving neural networks through augmenting topologies

An important question in neuroevolution is how to gain an advantage from evolving neural network topologies along with weights. We present a method, NeuroEvolution of Augmenting Topologies (NEAT), which outperforms the best fixed-topology method on a challenging benchmark reinforcement learning task. We claim that the increased efficiency is due to (1) employing a principled method of crossover of different topologies, (2) protecting structural innovation using speciation, and (3) incrementally growing from minimal structure. We test this claim through a series of ablation studies that demonstrate that each component is necessary to the system as a whole and to each other. What results is significantly faster learning. NEAT is also an important contribution to GAs because it shows how it is possible for evolution to both optimize and complexify solutions simultaneously, offering the possibility of evolving increasingly complex solutions over generations, and strengthening the analogy with biological evolution.     

Representation of neural networks through their multi-linearization

The studies on interpretability of neural networks have been playing an important role in understanding the knowledge developed through their learning and promoting the use of neurocomputing in practical problems. The rule-based setting in which neural networks are interpreted provides a convenient way of expressing knowledge in a transparent and modular manner and at a desired level of granularity (specificity). In this study, we formulate a certain engineering-based style of interpretation in which a given neural network is represented as a collection of local linear models where such models are developed around a collection of linearization nodes. The notion of multi-linearization of neural networks captures the essence of the proposed interpretation. We formulate the problem as an optimization of (i) a collection of linearization nodes around which individual linear models are formed and (ii) aggregation of the individual linearizations, where the linearization fields are subject to optimization. Given the non-differentiable character of the problem, we consider the use of population-based optimization of Particle Swarm Optimization (PSO). Numeric experiments are provided to illustrate the main aspects of the multi-linearization of neural networks.     

Dynamic neural cortical ensembles model

One-layered model of cortical neurons as a set of overlapping ensembles, each with a structure similar to Hopfield network, is proposed. Ensemble equilibrium equation is solved and formulas for connections weights calculation for given set of attractors are obtained. Concept of dynamic attractors that consists of consequent recalling of stored patterns with moving activity through the network is introduced. Role of dynamic attractors in long-term memory is discussed and mechanism for memory recovery after destruction of some neurons is proposed. Results of experiments on associative memory recovery after partial removal of neurons are shown.     

Evolving artificial neural network ensembles

Using a coordinated group of simple solvers to tackle a complex problem is not an entirely new idea. Its root could be traced back hundreds of years ago when ancient Chinese suggested a team approach to problem solving. For a long time, engineers have used the divide-and-conquer strategy to decompose a complex problem into simpler sub-problems and then solve them by a group of solvers. However, knowing the best way to divide a complex problem into simpler ones relies heavily on the available domain knowledge. It is often a manual process by an experienced engineer. There have been few automatic divide-and-conquer methods reported in the literature. Fortunately, evolutionary computation provides some of the interesting avenues to automatic divide-and-conquer methods. An in-depth study of such methods reveals that there is a deep underlying connection between evolutionary computation and ANN ensembles. Ideas in one area can be usefully transferred into another in producing effective algorithms. For example, using speciation to create and maintain diversity had inspired the development of negative correlation learning for ANN ensembles, and an in-depth study of diversity in ensembles. This paper will review some of the recent work in evolutionary approaches to designing ANN ensembles.     

Stationary ensembles in threshold networks

The networks of threshold elements are considered. The notion of a stationary ensemble is introduced. We set out necessary and sufficient conditions for a network to be an ensemble. We show that for two ensembles with common elements switching on one ensemble does not necessarily lead to switching on the other. A representation of an ensemble as a finite state machine is proposed. We then show how this representation can help study the processes of switching ensembles on and off. We note that a network of ensembles may be interpreted in neurobiological terms as a basis for a long-term memory model. In social sciences it may serve as a network version of the collective social threshold behavior model.     

Forming neural networks through efficient and adaptive coevolution

This article demonstrates the advantages of a cooperative, coevolutionary search in difficult control problems. The symbiotic adaptive neuroevolution (SANE) system coevolves a population of neurons that cooperate to form a functioning neural network. In this process, neurons assume different but overlapping roles, resulting in a robust encoding of control behavior. SANE is shown to be more efficient and more adaptive and to maintain higher levels of diversity than the more common network-based population approaches. Further empirical studies illustrate the emergent neuron specializations and the different roles the neurons assume in the population.     

小波和神经网络在心音识别中的应用

提出一种心音的特征提取和分类方法,用离散小波变换分解、重构产生信号的细节包络,进而用于提取特征,从预处理的信号中提取统计特性,作为心音分类的特征。多层感知器用于心音的分类,并通过250个心动周期得到验证,算法识别率达到92%。     

Improving optimization of convolutional neural networks through parameter fine-tuning

Neural Computing and Applications - In recent years, convolutional neural networks have achieved state-of-the-art performance in a number of computer vision problems such as image classification....     

Toward the scalability of neural networks through feature selection

In the past few years, the bottleneck for machine learning developers is not longer the limited data available but the algorithms inability to use all the data in the available time. For this reason, researches are now interested not only in the accuracy but also in the scalability of the machine learning algorithms. To deal with large-scale databases, feature selection can be helpful to reduce their dimensionality, turning an impracticable algorithm into a practical one. In this research, the influence of several feature selection methods on the scalability of four of the most well-known training algorithms for feedforward artificial neural networks (ANNs) will be analyzed over both classification and regression tasks. The results demonstrate that feature selection is an effective tool to improve scalability.     

Controlling industrial processes through supervised, feedforward neural networks

Multi-layer, feedforward perceptron neural networks produce hetero- and auto- pattern associators which can be applied to a wide range of problems. The area of process monitoring and control is one of numerous inputs and outputs, which are normally non-determinative and do not adhere to known probability distributions. By training neural networks through supervised learning, such as backpropagation, a mechanized tool can be created which offers advantages over traditional methods based on statistics. Significant benefits are the ability to discern complex relationships and trends rather than assuming distributions (usually Gaussian) or specifying algorithms, the ability to integrate in real time large amounts of continuous data and adapt incrementally to changes in process, and the ability to handle noisy or incomplete data.     

Dynamic analyses of information encoding in neural ensembles

Neural spike train decoding algorithms and techniques to compute Shannon mutual information are important methods for analyzing how neural systems represent biological signals. Decoding algorithms are also one of several strategies being used to design controls for brain-machine interfaces. Developing optimal strategies to design decoding algorithms and compute mutual information are therefore important problems in computational neuroscience. We present a general recursive filter decoding algorithm based on a point process model of individual neuron spiking activity and a linear stochastic state-space model of the biological signal. We derive from the algorithm new instantaneous estimates of the entropy, entropy rate, and the mutual information between the signal and the ensemble spiking activity. We assess the accuracy of the algorithm by computing, along with the decoding error, the true coverage probability of the approximate 0.95 confidence regions for the individual signal estimates. We illustrate the new algorithm by reanalyzing the position and ensemble neural spiking activity of CA1 hippocampal neurons from two rats foraging in an open circular environment. We compare the performance of this algorithm with a linear filter constructed by the widely used reverse correlation method. The median decoding error for Animal 1 (2) during 10 minutes of open foraging was 5.9 (5.5) cm, the median entropy was 6.9 (7.0) bits, the median information was 9.4 (9.4) bits, and the true coverage probability for 0.95 confidence regions was 0.67 (0.75) using 34 (32) neurons. These findings improve significantly on our previous results and suggest an integrated approach to dynamically reading neural codes, measuring their properties, and quantifying the accuracy with which encoded information is extracted.     

Perception control with improved expectation learning through multilayered neural networks

In this paper, we investigate the viability of multilayered neural network (NN)-based extension of a conventional "perception" control concept. The perception process selects and completes the information from the system to be controlled before passing it to the controlling agent so that control is not lost when sensory information from the system is incomplete. The perception process produces an expectation of the next set of information to be received from the system. The expectation is used to replace missing parts of the information received and it also influences the next perception. In the existing work, each of the expectation elements is linearly acquired such that the expectation tells only the dominant information in the recent past, i.e., this approach has no capability to sense the trend and the dynamics in the information. This handicap could become a serious problem when the perception process is applied to real physical systems. Here, we introduce an extension of the perception control process by using a radial basis function (RBF) feedforward NN to learn the trend and the dynamics in the information and produce the expectation of the next observation. Through some simulation comparisons, we show that the proposed RBFNN-based method is better than the existing one.     

Adaptive and intelligent wireless sensor networks through neural networks: an illustration for infrastructure adaptation through Hopfield network

This paper proposes embedding an artificial neural network into a wireless sensor network in fully parallel and distributed computation mode. The goal is to equip the wireless sensor network with computational intelligence and adaptation capability for enhanced autonomous operation. The applicability and utility of the proposed concept is demonstrated through a case study whereby a Hopfield neural network configured as a static optimizer for the weakly-connected dominating set problem is embedded into a wireless sensor network to enable it to adapt its network infrastructure to potential changes on-the-fly and following deployment in the field. Minimum weakly-connected dominating set defined for the graph model of the wireless sensor network topology is employed to represent the network infrastructure and can be recomputed each time the sensor network topology changes. A simulation study using the TOSSIM emulator for TinyOS-Mica sensor network platform was performed for mote counts of up to 1000. Time complexity, message complexity and solution quality measures were assessed and evaluated for the case study. Simulation results indicated that the wireless sensor network embedded with Hopfield neural network as a static optimizer performed competitively with other local or distributed algorithms for the weakly connected dominating set problem to establish its feasibility.     

Determining the number of real roots of polynomials through neural networks

The ability of feedforward neural networks to identify the number of real roots of univariate polynomials is investigated. Furthermore, their ability to determine whether a system of multivariate polynomial equations has real solutions is examined on a problem of determining the structure of a molecule. The obtained experimental results indicate that neural networks are capable of performing this task with high accuracy even when the training set is very small compared to the test set.     

Condition monitoring of 3G cellular networks through competitive neural models

We develop an unsupervised approach to condition monitoring of cellular networks using competitive neural algorithms. Training is carried out with state vectors representing the normal functioning of a simulated CDMA2000 network. Once training is completed, global and local normality profiles (NPs) are built from the distribution of quantization errors of the training state vectors and their components, respectively. The global NP is used to evaluate the overall condition of the cellular system. If abnormal behavior is detected, local NPs are used in a component-wise fashion to find abnormal state variables. Anomaly detection tests are performed via percentile-based confidence intervals computed over the global and local NPs. We compared the performance of four competitive algorithms [winner-take-all (WTA), frequency-sensitive competitive learning (FSCL), self-organizing map (SOM), and neural-gas algorithm (NGA)] and the results suggest that the joint use of global and local NPs is more efficient and more robust than current single-threshold methods.     

Optimization of neural networks through grammatical evolution and a genetic algorithm

This paper proposes a hybrid neuro-evolutive algorithm (NEA) that uses a compact indirect encoding scheme (IES) for representing its genotypes (a set of ten production rules of a Lindenmayer System with memory), moreover has the ability to reuse the genotypes and automatically build modular, hierarchical and recurrent neural networks. A genetic algorithm (GA) evolves a Lindenmayer System (L-System) that is used to design the neural network’s architecture. This basic neural codification confers scalability and search space reduction in relation to other methods. Furthermore, the system uses a parallel genome scan engine that increases both the implicit parallelism and convergence of the GA. The fitness function of the NEA rewards economical artificial neural networks (ANNs) that are easily implemented. The NEA was tested on five real-world classification datasets and three well-known datasets for time series forecasting (TSF). The results are statistically compared against established state-of-the-art algorithms and various forecasting methods (ADANN, ARIMA, UCM, and Forecast Pro). In most cases, our NEA outperformed the other methods, delivering the most accurate classification and time series forecasting with the least computational effort. These superior results are attributed to the improved effectiveness and efficiency of NEA in the decision-making process. The result is an optimized neural network architecture for solving classification problems and simulating dynamical systems.     

Transfer of radiative heat through clothing ensembles

A mathematical model was designed to calculate the temperature and dry heat transfer in the various layers of a clothing ensemble, and the total heat loss of a human who is irradiated for a certain fraction of his or her area. The clothing ensemble that is irradiated by an external heat source is considered to be composed of underclothing, trapped air, and outer fabric. The model was experimentally tested with heat balance methods, using subjects, varying the activity, wind, and radiation characteristics of the outer garment of two-layer ensembles. In two experiments the subjects could only give off dry heat because they were wrapped in plastic foil. The model appeared to be correct within about l°C (rms error) and l0Wm^?2 (rms error). In a third experiment, sweat evaporation was also taken into account, showing that the resulting physiological heat load of 10 to 30% of the intercepted additional radiation is compensated by additional sweating. The resulting heat strain was rather mild. It is concluded that the mathematical model is a valid tool for the investigation of heat transfer through two-layer ensembles in radiant environments.