A DATA MINING METHOD BASED ON CONSTRUCTIVE NEURAL NETWORKS

In this letter,Constructive Neural Networks (CNN) is used in large-scale data mining. By introducing the principle and characteristics of CNN and pointing out its deficiencies,fuzzy theory is adopted to improve the covering algorithms. The threshold of covering algorithms is redefined. "Extended area" for test samples is built. The inference of the outlier is eliminated. Furthermore,"Sphere Neighborhood (SN)" are constructed. The membership functions of test samples are given and all of the test samples are determined accordingly. The method is used to mine large wireless monitor data (about 3×107 data points),and knowledge is found effectively.

Indian stock market prediction using artificial neural networks on tick data

Introduction:Nowadays,the most significant challenges in the stock market is to predict the stock prices.The stock price data represents a financial time series data which becomes more difficult to predict due to its characteristics and dynamic nature.Case description:Support Vector Machines(SVM)and Artificial Neural Networks(ANN)are widely used for prediction of stock prices and its movements.Every algorithm has its way of learning patterns and then predicting.Artificial Neural Network(ANN)is a popular method which also incorporate technical analysis for making predictions in financial markets.Discussion and evaluation:Most common techniques used in the forecasting of financial time series are Support Vector Machine(SVM),Support Vector Regression(SVR)and Back Propagation Neural Network(BPNN).In this article,we use neural networks based on three different learning algorithms,i.e.,Levenberg-Marquardt,Scaled Conjugate Gradient and Bayesian Regularization for stock market prediction based on tick data as well as 15-min data of an Indian company and their results compared.Conclusion:All three algorithms provide an accuracy of 99.9%using tick data.The accuracy over 15-min dataset drops to 96.2%,97.0%and 98.9%for LM,SCG and Bayesian Regularization respectively which is significantly poor in comparison with that of results obtained using tick data.

An Optimized Data Fusion Paradigm for WSN Based on Neural Networks

Wireless sensor networks(WSNs)have gotten a lot of attention as useful tools for gathering data.The energy problem has been a fundamental constraint and challenge faced by many WSN applications due to the size and cost constraints of the sensor nodes.This paper proposed a data fusion model based on the back propagation neural network(BPNN)model to address the problem of a large number of invalid or redundant data.Using three layeredbased BPNNs and a TEEN threshold,the proposed model describes the cluster structure and filters out unnecessary details.During the information transmission process,the neural network’s output function is used to deal with a large amount of sensing data,where the feature value of sensing data is extracted and transmitted to the sink node.In terms of life cycle,data traffic,and network use,simulation results show that the proposed data fusion model outperforms the traditional TEEN protocol.As a result,the proposed scheme increases the life cycle of the network thereby lowering energy usage and traffic.

A New Class of Nonlinear Error Control Codes Based on Neural Networks

By mcans of stable attractors of discret Hopfield neural network (DHNN) , anew class of nonlinear error control codes is sugsested and some relativetheorems are presented. A kind of single error control codes is also given forillustrating this new approach.

Stochastic asymptotical synchronization of chaotic Markovian jumping fuzzy cellular neural networks with mixed delays and the Wiener process based on sampled-data control

We investigate the stochastic asymptotical synchronization of chaotic Markovian jumping fuzzy cellular neural networks （MJFCNNs） with discrete, unbounded distributed delays, and the Wiener process based on sampled-data control using the linear matrix inequality （LMI） approach. The Lyapunov–Krasovskii functional combined with the input delay approach as well as the free-weighting matrix approach is employed to derive several sufficient criteria in terms of LMIs to ensure that the delayed MJFCNNs with the Wiener process is stochastic asymptotical synchronous. Restrictions （e.g., time derivative is smaller than one） are removed to obtain a proposed sampled-data controller. Finally, a numerical example is provided to demonstrate the reliability of the derived results.

Discovering Frequent Subtrees from XML Data Using Neural Networks

By rapid progress of network and storage technologies, a huge amount of electronic data such as Web pages and XML has been available on Internet. In this paper, we study a data-mining problem of discovering frequent ordered sub-trees in a large collection of XML data, where both of the patterns and the data are modeled by labeled ordered trees. We present an efficient algorithm of Ordered Subtree Miner （OSTMiner） based on two- layer neural networks with Hebb rule, that computes all ordered sub-trees appearing in a collection of XML trees with frequent above a user-specified threshold using a special structure EM-tree. In this algo- rithm, EM-tree is used as an extended merging tree to supply scheme information for efficient pruning and mining frequent sub-trees. Experiments results showed that OSTMiner has good response time and scales well.

Free-matrix-based time-dependent discontinuous Lyapunov functional for synchronization of delayed neural networks with sampled-data control

This paper is concerned with the synchronization of delayed neural networks via sampled-data control. A new technique, namely, the free-matrix-based time-dependent discontinuous Lyapunov functional approach, is adopted in constructing the Lyapunov functional, which takes advantage of the sampling characteristic of sawtooth input delay. Based on this discontinuous Lyapunov functional, some less conservative synchronization criteria are established to ensure that the slave system is synchronous with the master system. The desired sampled-data controller can be obtained through the use of the linear matrix inequality（LMI） technique. Finally, two numerical examples are provided to demonstrate the effectiveness and the improvements of the proposed methods.

Stock Price Prediction Using Predictive Error Compensation Wavelet Neural Networks

:Machine Learning(ML)algorithms have been widely used for financial time series prediction and trading through bots.In this work,we propose a Predictive Error Compensated Wavelet Neural Network(PEC-WNN)ML model that improves the prediction of next day closing prices.In the proposed model we use multiple neural networks where the first one uses the closing stock prices from multiple-scale time-domain inputs.An additional network is used for error estimation to compensate and reduce the prediction error of the main network instead of using recurrence.The performance of the proposed model is evaluated using six different stock data samples in the New York stock exchange.The results have demonstrated significant improvement in forecasting accuracy in all cases when the second network is used in accordance with the first one by adding the outputs.The RMSE error is 33%improved when the proposed PEC-WNN model is used compared to the Long ShortTerm Memory(LSTM)model.Furthermore,through the analysis of training mechanisms,we found that using the updated training the performance of the proposed model is improved.The contribution of this study is the applicability of simultaneously different time frames as inputs.Cascading the predictive error compensation not only reduces the error rate but also helps in avoiding overfitting problems.

Bitcoin Candlestick Prediction with Deep Neural Networks Based on Real Time Data

Currently,Bitcoin is the world’s most popular cryptocurrency.The price of Bitcoin is extremely volatile,which can be described as high-benefit and high-risk.To minimize the risk involved,a means of more accurately predicting the Bitcoin price is required.Most of the existing studies of Bitcoin prediction are based on historical(i.e.,benchmark)data,without considering the real-time(i.e.,live)data.To mitigate the issue of price volatility and achieve more precise outcomes,this study suggests using historical and real-time data to predict the Bitcoin candlestick—or open,high,low,and close(OHLC)—prices.Seeking a better prediction model,the present study proposes time series-based deep learning models.In particular,two deep learning algorithms were applied,namely,long short-term memory(LSTM)and gated recurrent unit(GRU).Using real-time data,the Bitcoin candlesticks were predicted for three intervals:the next 4 h,the next 12 h,and the next 24 h.The results showed that the best-performing model was the LSTM-based model with the 4-h interval.In particular,this model achieved a stellar performance with a mean absolute percentage error(MAPE)of 0.63,a root mean square error(RMSE)of 0.0009,a mean square error(MSE)of 9e-07,a mean absolute error(MAE)of 0.0005,and an R-squared coefficient(R2)of 0.994.With these results,the proposed prediction model has demonstrated its efficiency over the models proposed in previous studies.The findings of this study have considerable implications in the business field,as the proposed model can assist investors and traders in precisely identifying Bitcoin sales and buying opportunities.

Alternative Method of Constructing Granular Neural Networks

Utilizing granular computing to enhance artificial neural network architecture, a newtype of network emerges—thegranular neural network (GNN). GNNs offer distinct advantages over their traditional counterparts: The ability toprocess both numerical and granular data, leading to improved interpretability. This paper proposes a novel designmethod for constructing GNNs, drawing inspiration from existing interval-valued neural networks built uponNNNs. However, unlike the proposed algorithm in this work, which employs interval values or triangular fuzzynumbers for connections, existing methods rely on a pre-defined numerical network. This new method utilizesa uniform distribution of information granularity to granulate connections with unknown parameters, resultingin independent GNN structures. To quantify the granularity output of the network, the product of two commonperformance indices is adopted: The coverage of numerical data and the specificity of information granules.Optimizing this combined performance index helps determine the optimal parameters for the network. Finally,the paper presents the complete model construction and validates its feasibility through experiments on datasetsfrom the UCIMachine Learning Repository. The results demonstrate the proposed algorithm’s effectiveness andpromising performance.

Pattern recognition and data mining software based on artificial neural networks applied to proton transfer in aqueous environments

In computational physics proton transfer phenomena could be viewed as pattern classification problems based on a set of input features allowing classification of the proton motion into two categories： transfer ＇occurred＇ and transfer ＇not occurred＇. The goal of this paper is to evaluate the use of artificial neural networks in the classification of proton transfer events, based on the feed-forward back propagation neural network, used as a classifier to distinguish between the two transfer cases. In this paper, we use a new developed data mining and pattern recognition tool for automating, controlling, and drawing charts of the output data of an Empirical Valence Bond existing code. The study analyzes the need for pattern recognition in aqueous proton transfer processes and how the learning approach in error back propagation （multilayer perceptron algorithms） could be satisfactorily employed in the present case. We present a tool for pattern recognition and validate the code including a real physical case study. The results of applying the artificial neural networks methodology to crowd patterns based upon selected physical properties （e.g., temperature, density） show the abilities of the network to learn proton transfer patterns corresponding to properties of the aqueous environments, which is in turn proved to be fully compatible with previous proton transfer studies.

Effective data sampling strategies and boundary condition constraints of physics-informed neural networks for identifying material properties in solid mechanics

Material identification is critical for understanding the relationship between mechanical properties and the associated mechanical functions.However,material identification is a challenging task,especially when the characteristic of the material is highly nonlinear in nature,as is common in biological tissue.In this work,we identify unknown material properties in continuum solid mechanics via physics-informed neural networks(PINNs).To improve the accuracy and efficiency of PINNs,we develop efficient strategies to nonuniformly sample observational data.We also investigate different approaches to enforce Dirichlet-type boundary conditions(BCs)as soft or hard constraints.Finally,we apply the proposed methods to a diverse set of time-dependent and time-independent solid mechanic examples that span linear elastic and hyperelastic material space.The estimated material parameters achieve relative errors of less than 1%.As such,this work is relevant to diverse applications,including optimizing structural integrity and developing novel materials.

A Hybrid Method for Compression of Solar Radiation Data Using Neural Networks

The prediction of solar radiation is important for several applications in renewable energy research. There are a number of geographical variables which affect solar radiation prediction, the identification of these variables for accurate solar radiation prediction is very important. This paper presents a hybrid method for the compression of solar radiation using predictive analysis. The prediction of minute wise solar radiation is performed by using different models of Artificial Neural Networks (ANN), namely Multi-layer perceptron neural network (MLPNN), Cascade feed forward back propagation (CFNN) and Elman back propagation (ELMNN). Root mean square error (RMSE) is used to evaluate the prediction accuracy of the three ANN models used. The information and knowledge gained from the present study could improve the accuracy of analysis concerning climate studies and help in congestion control.

Artificial Neural Networks Application to Predict Wheat Yield Using Climatic Data

The goal of this study was to apply artificial neural networks to predict rain-fed wheat yield using meteorological data a few days to few months before harvesting. The climatic observation data used; were mean of daily minimum and maximum temperature, extreme of daily minimum and maximum temperature, sum of daily rainfall, number of rainy days, sum of daily sun hours, mean of daily wind speed, extreme of daily wind speed, mean of daily relative humidity, and sum of daily water requirements that were collected during 1990-1999 in Sararood Station for wheat phenological stages consisting; sowing, germination, emergence, 3rd leaves, tillering, stem formation, heading, flowering, milk maturity, wax maturity, full maturity, separately for each growing season. Then, they arranged in a matrix whose rows form each of the statistical years and the columns are meteorological factors at each phenological stage. Finally, the obtained model had the following capabilities： Prediction of wheat yield with maximum errors of 45-60 kg/ha at least two months before full maturity stage, determination of the sensitivity of each phenological stage with respect to meteorological factors, and determination of the priority order and importance of each meteorological factor effective in plant growth and crop yield.

A Novel Approach to Disqualify Datasets Using Accumulative Statistical Spread Map with Neural Networks (ASSM-NN)

A novel approach to detect and filter out an unhealthy dataset from a matrix of datasets is developed, tested, and proved. The technique employs a new type of self organizing map called Accumulative Statistical Spread Map (ASSM) to establish the destructive and negative effect a dataset will have on the rest of the matrix if stayed within that matrix. The ASSM is supported by training a neural network engine, which will determine which dataset is responsible for its inability to learn, classify and predict. The carried out experiments proved that a neural system was not able to learn in the presence of such an unhealthy dataset that possessed some deviated characteristics, even though it was produced under the same conditions and through the same process as the rest of the datasets in the matrix, and hence, it should be disqualified, and either removed completely or transferred to another matrix. Such novel approach is very useful in pattern recognition of datasets and features that do not belong to their source and could be used as an effective tool to detect suspicious activities in many areas of secure filing, communication and data storage.

Convolutional neural networks for time series classification

Time series classification is an important task in time series data mining, and has attracted great interests and tremendous efforts during last decades. However, it remains a challenging problem due to the nature of time series data: high dimensionality, large in data size and updating continuously. The deep learning techniques are explored to improve the performance of traditional feature-based approaches. Specifically, a novel convolutional neural network (CNN) framework is proposed for time series classification. Different from other feature-based classification approaches, CNN can discover and extract the suitable internal structure to generate deep features of the input time series automatically by using convolution and pooling operations. Two groups of experiments are conducted on simulated data sets and eight groups of experiments are conducted on real-world data sets from different application domains. The final experimental results show that the proposed method outperforms state-of-the-art methods for time series classification in terms of the classification accuracy and noise tolerance. © 1990-2011 Beijing Institute of Aerospace Information.

Synchronization of chaos using radial basis functions neural networks

The Radial Basis Functions Neural Network （RBFNN） is used to establish the model of a response system through the input and output data of the system. The synchronization between a drive system and the response system can be implemented by employing the RBFNN model and state feedback control. In this case, the exact mathematical model, which is the precondition for the conventional method, is unnecessary for implementing synchronization. The effect of the model error is investigated and a corresponding theorem is developed. The effect of the parameter perturbations and the measurement noise is investigated through simulations. The simulation results under different conditions show the effectiveness of the method.

Topological probability and connection strength induced activity in complex neural networks

Recent experimental evidence suggests that some brain activities can be assigned to small-world networks. In this work, we investigate how the topological probability p and connection strength C affect the activities of discrete neural networks with small-world （SW） connections. Network elements are described by two-dimensional map neurons （2DMNs） with the values of parameters at which no activity occurs. It is found that when the value of p is smaller or larger, there are no active neurons in the network, no matter what the value of connection strength is; for a given appropriate connection strength, there is an intermediate range of topological probability where the activity of 2DMN network is induced and enhanced. On the other hand, for a given intermediate topological probability level, there exists an optimal value of connection strength such that the frequency of activity reaches its maximum. The possible mechanism behind the action of topological probability and connection strength is addressed based on the bifurcation method. Furthermore, the effects of noise and transmission delay on the activity of neural network are also studied.

Modeling of Trophospheric Ozone Concentrations Using Genetically Trained Multi-Level Cellular Neural Networks

Tropospheric ozone concentrations, which are an important air pollutant, are modeled by the use of an artificial intelligence structure. Data obtained from air pollution measurement stations in the city of Istanbul are utilized in constituting the model. A supervised algorithm for the evaluation of ozone concentration using a genetically trained multi-level cellular neural network （ML-CNN） is introduced, developed, and applied to real data. A genetic algorithm is used in the optimization of CNN templates. The model results and the actual measurement results are compared and statistically evaluated. It is observed that seasonal changes in ozone concentrations are reflected effectively by the concentrations estimated by the multilevel-CNN model structure, with a correlation value of 0.57 ascertained between actual and model results. It is shown that the multilevel-CNN modeling technique is as satisfactory as other modeling techniques in associating the data in a complex medium in air pollution applications.