Neural network ensemble operators for time series forecasting

The combination of forecasts resulting from an ensemble of neural networks has been shown to outperform the use of a single “best” network model. This is supported by an extensive body of literature, which shows that combining generally leads to improvements in forecasting accuracy and robustness, and that using the mean operator often outperforms more complex methods of combining forecasts. This paper proposes a mode ensemble operator based on kernel density estimation, which unlike the mean operator is insensitive to outliers and deviations from normality, and unlike the median operator does not require symmetric distributions. The three operators are compared empirically and the proposed mode ensemble operator is found to produce the most accurate forecasts, followed by the median, while the mean has relatively poor performance. The findings suggest that the mode operator should be considered as an alternative to the mean and median operators in forecasting applications. Experiments indicate that mode ensembles are useful in automating neural network models across a large number of time series, overcoming issues of uncertainty associated with data sampling, the stochasticity of neural network training, and the distribution of the forecasts.     

Improving artificial neural networks’ performance in seasonal time series forecasting

In this study, an artificial neural network (ANN) structure is proposed for seasonal time series forecasting. The proposed structure considers the seasonal period in time series in order to determine the number of input and output neurons. The model was tested for four real-world time series. The results found by the proposed ANN were compared with the results of traditional statistical models and other ANN architectures. This comparison shows that the proposed model comes with lower prediction error than other methods. It is shown that the proposed model is especially convenient when the seasonality in time series is strong; however, if the seasonality is weak, different network structures may be more suitable.     

基于混沌与改进BP神经网络的电价预测方法

结合混沌分析理论和BP神经网络,提出在混沌相空间建立BP神经网络模型.对美国加州边际电价进行预测,并对预测结果进行分析,取得了满意的结果.     

Artificial neural networks: a review of commercial hardware

Artificial neural networks (ANN) became a common solution for a wide variety of problems in many fields, such as control and pattern recognition to name but a few. Many solutions found in these and other ANN fields have reached a hardware implementation phase, either commercial or with prototypes. The most frequent solution for the implementation of ANN consists of training and implementing the ANN within a computer. Nevertheless this solution might be unsuitable because of its cost or its limited speed. The implementation might be too expensive because of the computer and too slow when implemented in software. In both cases dedicated hardware can be an interesting solution.

The necessity of dedicated hardware might not imply building the hardware since in the last two decades several commercial hardware solutions that can be used in the implementation have reached the market.

Unfortunately not every integrated circuit will fit the needs: some will use lower precision, some will implement only certain types of networks, some don’t have training built in and the information is not easy to find.

This article is confined to reporting the commercial chips that have been developed specifically for ANN, leaving out other solutions.

This option has been made because most of the other solutions are based on cards which are built either with these chips, Digital Signal Processors or Reduced Instruction Set Computers.     

Review of pulse-coupled neural networks

This paper reviews the research status of pulse-coupled neural networks (PCNN) in the past decade. Considering there are too many publications about the PCNN, we summarize main approaches and point out interesting parts of the PCNN researches rather than contemplate to go into details of particular algorithms or describe results of comparative experiments. First, the current status of the PCNN and some modified models are briefly introduced. Second, we review the PCNN applications in the field of image processing (e.g. image segmentation, image enhancement, image fusion, object and edge detection, pattern recognition, etc.), then applications in other fields also are mentioned. Subsequently, some existing problems are summarized, while we give some suggestions for the solutions to some puzzles. Finally, the trend of the PCNN is pointed out.     

基于神经网络集成的汽车牌照识别

对基于神经网络集成的汽车牌照识别的原理和方法进行了研究,并着重分析了现有技术的积极因素和潜在问题,提出了一种基于神经网络集成进行车牌文字识别的方法.在特征提取时采用了多种特征提取的方法,对提取的每种特征构建一个BP神经网络分别进行训练.最终待识别的字符将被神经网络集成进行识别.实践证明,利用该方法比单个神经网络识别有更高的识别率,具有较高的使用价值.     

An investigation and comparison of artificial neural network and time series models for Chinese food grain price forecasting

This paper compares the predictive performance of ARIMA, artificial neural network and the linear combination models for forecasting wheat price in Chinese market. Empirical results show that the combined model can improve the forecasting performance significantly in contrast with its counterparts in terms of the error evaluation measurements. However, as far as turning points and profit criterions are concerned, the ANN model is best as well as at capturing a significant number of turning points. The results are conflicting when implementing dissimilar forecasting criteria (the quantitative and the turning points measurements) to evaluate the performance of three models. The ANN model is overall the best model, and can be used as an alternative method to model Chinese future food grain price.     

Classification error of multilayer perceptron neural networks

In subject classification, artificial neural networks (ANNS) are efficient and objective classification methods. Thus, they have been successfully applied to the numerous classification fields. Sometimes, however, classifications do not match the real world, and are subjected to errors. These problems are caused by the nature of ANNS. We discuss these on multilayer perceptron neural networks. By studying of these problems, it helps us to have a better understanding on its classification.     

基于VLRBP神经网络的汇率预测

为了提高汇率预测的准确性,分别使用VLRBP神经网络模型和GRNN模型及ARIMA模型对欧元汇率时间序列进行建模和预测,通过实证分析发现基于VLRBP的神经网络对于含有大量非线性成分的欧元汇率时间序列的预测比较准确.在分析了最速下降BP学习算法的缺点后,提出利用VLRBP学习算法来解决神经网络振荡和收敛速度过慢的缺陷,并取得较好的效果.同时,为了提高VLRBP网络的泛化性能,提出在训练VLRBP神经网络时应用浴盆曲线方法选取隐层神经元个数和滑动窗口尺寸,试验结果表明该方法适合神经网络模型.     

A high order fuzzy time series forecasting model based on adaptive expectation and artificial neural networks

Many fuzzy time series approaches have been proposed in recent years. These methods include three main phases such as fuzzification, defining fuzzy relationships and, defuzzification. Aladag et al. [2] improved the forecasting accuracy by utilizing feed forward neural networks to determine fuzzy relationships in high order fuzzy time series. Another study for increasing forecasting accuracy was made by Cheng et al. [6]. In their study, they employ adaptive expectation model to adopt forecasts obtained from first order fuzzy time series forecasting model. In this study, we propose a novel high order fuzzy time series method in order to obtain more accurate forecasts. In the proposed method, fuzzy relationships are defined by feed forward neural networks and adaptive expectation model is used for adjusting forecasted values. Unlike the papers of Cheng et al. [6] and Liu et al. [14], forecast adjusting is done by using constraint optimization for weighted parameter. The proposed method is applied to the enrollments of the University of Alabama and the obtained forecasting results compared to those obtained from other approaches are available in the literature. As a result of comparison, it is clearly seen that the proposed method significantly increases the forecasting accuracy.     

An application of pruning in the design of neural networks for real time flood forecasting

We propose the application of pruning in the design of neural networks for hydrological prediction. The basic idea of pruning algorithms, which have not been used in water resources problems yet, is to start from a network which is larger than necessary, and then remove the parameters that are less influential one at a time, designing a much more parameter-parsimonious model. We compare pruned and complete predictors on two quite different Italian catchments. Remarkably, pruned models may provide better generalization than fully connected ones, thus improving the quality of the forecast. Besides the performance issues, pruning is useful to provide evidence of inputs relevance, removing measuring stations identified as redundant (30–40% in our case studies) from the input set. This is a desirable property in the system exercise since data may not be available in extreme situations such as floods; the smaller the set of measuring stations the model depends on, the lower the probability of system downtimes due to missing data. Furthermore, the Authority in charge of the forecast system may decide for real-time operations just to link the gauges of the pruned predictor, thus saving costs considerably, a critical issue in developing countries.

基于径向基函数(RBF)的安徽省GDP增长模拟与预测

本文运用新型非线性径向基函数RBF神经网络模型，对安徽省国内生产总值(GDP)进行了宏观经济模拟预测分析，结果证明与其它经济计量方法相比较，网络模型新颖，具有较好的预测精度及效果，可广泛应用于各种预测研究，有较高的应用推广价值。     

Wavelet-based multi-resolution analysis and artificial neural networks for forecasting temperature and thermal power consumption

As part of the OptiEnR research project, the present paper deals with outdoor temperature and thermal power consumption forecasting. This project focuses on optimizing the functioning of a multi-energy district boiler (La Rochelle, west coast of France), adding to the plant a thermal storage unit and implementing a model-based predictive controller. The proposed short-term forecast method is based on the concept of time series and uses both a wavelet-based multi-resolution analysis and multi-layer artificial neural networks. One could speak of “MRA-ANN” methodology. The discrete wavelet transform allows decomposing sequences of past data in subsequences (named coefficients) according to different frequency domains, while preserving their temporal characteristics. From these coefficients, multi-layer Perceptrons are used to estimate future subsequences of 4 h and 30 min. Future values of outdoor temperature and thermal power consumption are then obtained by simply summing up the estimated coefficients. Substituting the prediction task of an original time series of high variability with the estimation of its wavelet coefficients on different levels of lower variability is the main idea of the present work. In addition, the sequences of past data are completed, for each of their components, by both the minute of the day and the day of the year to place the developed model in time. The present paper mainly focuses on the impact on forecast accuracy of various parameters, related with the discrete wavelet transform, such as both the wavelet order and the decomposition level, and the topology of the neural networks used. The number of past sequences to take into account and the chosen time step were also major concerns. The optimal configuration for the tools used leads to very good forecasting results and validates the proposed MRA-ANN methodology.     

TACO-miner: An ant colony based algorithm for rule extraction from trained neural networks

Extracting classification rules from data is an important task of data mining and gaining considerable more attention in recent years. In this paper, a new meta-heuristic algorithm which is called as TACO-miner is proposed for rule extraction from artificial neural networks (ANN). The proposed rule extraction algorithm actually works on the trained ANNs in order to discover the hidden knowledge which is available in the form of connection weights within ANN structure. The proposed algorithm is mainly based on a meta-heuristic which is known as touring ant colony optimization (TACO) and consists of two-step hierarchical structure. The proposed algorithm is experimentally evaluated on six binary and n-ary classification benchmark data sets. Results of the comparative study show that TACO-miner is able to discover accurate and concise classification rules.     

A comparative study for forecasting using neural networks vs genetically identified Box&Jenkins models

This paper aims to discuss the results and conclusions of an extensive comparative study on the forecasting performance between two different techniques: a genetic expert system in which a genetic algorithm carries out the identification stage embraced in the three- phase Box&Jenkins univariate methodology; and a connectionist approach. At the heart of the former, an expert system rules the identification-estimation-diagnostic checking cyclical process to end up with the predictions provided by the SARIMA model which best fits the data. We will present the connectionist approach as technically equivalent to the latter process and due to its, alas, lack of any conclusive existent algorithm able to identify both the optimal model and architecture for a given problem, the three most common models presently at use and 20 different architectures for each model will be examined. It seems natural that if a comparison is to be made in order to provide a straight answer as to whether or not a connectionist approach outperforms the univariate Box&Jenkins methodology, the benchmark should clearly be the set of time series analysed in the work Time Series Analysis. Forecasting and Control by G. E. Box and G. M. Jenkins. Series BJA through to BJG give a total of 1200 plus measures to evaluate and compare the predictive power for different models, architectures, prediction horizons and pre-processing transformations.     

Diagnosing diabetes using neural networks on small mobile devices

Pervasive computing is often mentioned in the context of improving healthcare. This paper presents a novel approach for diagnosing diabetes using neural networks and pervasive healthcare computing technologies. The recent developments in small mobile devices and wireless communications provide a strong motivation to develop new software techniques and mobile services for pervasive healthcare computing. A distributed end-to-end pervasive healthcare system utilizing neural network computations for diagnosing illnesses was developed. This work presents the initial results for a simple client (patient’s PDA) and server (powerful desktop PC) two-tier pervasive healthcare architecture. The computations of neural network operations on both client and server sides and wireless network communications between them are optimized for real time use of pervasive healthcare services.     

A flexible neural network-fuzzy mathematical programming algorithm for improvement of oil price estimation and forecasting

This paper presents a flexible algorithm based on artificial neural network (ANN) and fuzzy regression (FR) to cope with optimum long-term oil price forecasting in noisy, uncertain, and complex environments. The oil supply, crude oil distillation capacity, oil consumption of non-OECD, USA refinery capacity, and surplus capacity are incorporated as the economic indicators. Analysis of variance (ANOVA) and Duncan’s multiple range test (DMRT) are then applied to test the significance of the forecasts obtained from ANN and FR models. It is concluded that the selected ANN models considerably outperform the FR models in terms of mean absolute percentage error (MAPE). Moreover, Spearman correlation test is applied for verification and validation of the results. The proposed flexible ANN–FR algorithm may be easily modified to be applied to other complex, non-linear and uncertain datasets.     

Optimized scenario for rainfall forecasting using genetic algorithm coupled with artificial neural network

Rainfall forecasting plays many important role in water resources studies such as river training works and design of flood warning systems. Recent advancement in artificial intelligence and in particular techniques aimed at converting input to output for highly nonlinear, non-convex and dimensionalized processes such as rainfall field, provide an alternative approach for developing rainfall forecasting model. Artificial neural networks (ANNs), which perform a nonlinear mapping between inputs and outputs, are such a technique. Current literatures on artificial neural networks show that the selection of network architecture and its efficient training procedure are major obstacles for their daily usage. In this paper, feed-forward type networks will be developed to simulate the rainfall field and a so-called back propagation (BP) algorithm coupled with genetic algorithm (GA) will be used to train and optimize the networks. The technique will be implemented to forecast rainfall for a number of times using rainfall hyetograph of recording rain gauges in the Upper Parramatta catchment in the western suburbs of Sydney, Australia. Results of the study showed the structuring of ANN network with the input parameter selection, when coupled with GA, performed better compared to similar work of using ANN alone.     

Fast evaluation of neural networks via confidence rating

Neural networks have become very useful tools for input–output knowledge discovery. However, some of the most powerful schemes require very complex machines and, thus, a large amount of calculation. This paper presents a general technique to reduce the computational burden associated with the operational phase of most neural networks that calculate their output as a weighted sum of terms, which comprises a wide variety of schemes, such as Multi-Net or Radial Basis Function networks. Basically, the idea consists on sequentially evaluating the sum terms, using a series of thresholds which are associated with the confidence that a partial output will coincide with the overall network classification criterion. Furthermore, we design some procedures for conveniently sorting out the network units, so that the most important ones are evaluated first. The possibilities of this strategy are illustrated with some experiments on a benchmark of binary classification problems, using RealAdaboost and RBF networks, which show that important computational savings can be achieved without significant degradation in terms of recognition accuracy.     

Feedforward backpropagation artificial neural networks on reconfigurable meshes

The artificial neural networks (ANNs) have been used successfully in applications such as pattern recognition, image processing, automation and control. Majority of today's applications use backpropagate feedforward ANN. In this paper, two methods of P pattern L layer ANN learning on n × n RMESH have been presented. One required memory space of O(nL) but conceptually is simpler to develop and the other uses pipelined approach which reduces the memory requirement to O(L). Both of these algorithms take O(PL) time and are optimal for RMESH architecture.