Qingfang Meng

Small-time scale network traffic prediction based on flexible neural tree

In this paper, the flexible neural tree (FNT) model is employed to predict the small-time scale traffic measurements data. Based on the pre-defined instruction/operator sets, the FNT model can be created and evolved. This framework allows input variables selection, over-layer connections and different activation functions for the various nodes involved. The FNT structure is developed using the Genetic Programming (GP) and the parameters are optimized by the Particle Swarm Optimization algorithm (PSO). The experimental results indicate that the proposed method is efficient for forecasting small-time scale traffic measurements and can reproduce the statistical features of real traffic measurements. We also compare the performance of the FNT model with the feed-forward neural network optimized by PSO for the same problem.

Time-series forecasting using a system of ordinary differential equations

This paper presents a hybrid evolutionary method for identifying a system of ordinary differential equations (ODEs) to predict the small-time scale traffic measurements data. We used the tree-structure based evolutionary algorithm to evolve the architecture and a particle swarm optimization (PSO) algorithm to fine tune the parameters of the additive tree models for the system of ordinary differential equations. We also illustrate some experimental comparisons with genetic programming, gene expression programming and a feedforward neural network optimized using PSO algorithm. Experimental results reveal that the proposed method is feasible and efficient for forecasting the small-scale traffic measurements data.

Classification of Ventricular Tachycardia and Fibrillation Based on the Lempel-Ziv Complexity and EMD

Detection of ventricular tachycardia (VT) and ventricular fibrillation (VF) in electrocardiography (ECG) has clinical research significance. The complexity of the heart signals has changed significantly, when the heart state switches from normal sinus rhythm to VT or VF. With the consideration of the non-stationary of VT and VF, we proposed a novel method for classification of VF and VT in this paper, based on the Lempel-Ziv (LZ) complexity and empirical mode decomposition (EMD). The EMD first decomposed ECG signals into a set of intrinsic mode functions (IMFs). Then the complexity of each IMF was used as a feature in order to discriminate between VF and VT. A public dataset was utilized for evaluating the proposed method. Experimental results showed that the proposed method could successfully distinguish VF from VT with the highest accuracy up to 97.08%.

The feature extraction method of EEG signals based on degree distribution of complex networks from nonlinear time series

The nonlinear time series analysis method based on complex networks theory gives a novel perspective to understand the dynamics of the nonlinear time series. Considering the electroencephalogram (EEG) signals showing different nonlinear dynamics under different brain states, this study proposes an epileptic EEG analysis approach based on statistical properties of complex networks and applies the approach to epileptic EEGs automatic detection. Firstly, the complex network is constructed from the epileptic EEG signals and the degree distribution (DDF) of the resulting networks is calculated. Then the entropy of the degree distribution (NDDE) is used as a feature to classify the ictal EEGs and the interictal EEGs. The experiment results show that the NDDE of the ictal EEG is lower than interictal EEGs and the classification accuracy, taking the NDDE as a classification feature, is up to 96.25%.

Seizure detection in clinical EEG based on entropies and EMD

Considering the EEG signals are nonlinear and nonstationary, the nonlinear dynamical methods have been widely applied to analyze the EEG signals. Directly extracted the approximate entropy and sample entropy as features are efficient methods to analysis the EEG signals of epileptic parents. To detect the epilepsy seizure signals from epileptic EEG, choose an appropriate threshold value as the discrimination criteria is simplest. The experiment indicated the approximate entropy provide a higher accuracy in distinguishing the epileptic seizure signals from the EEG than sample entropy. To improve the accuracy of sample entropy, empirical mode decomposition (EMD) is used to decompose EEG into multiple frequency subbands, and then calculate sample entropy for each component. The results show that the accuracy is up to 91%, which could be used to discriminate epileptic seizure signals from epileptic EEG.

Ensemble of Flexible Neural Tree and Ordinary Differential Equations for Small-Time Scale Network Traffic Prediction

Accurate models play important roles in capturing the salient characteristics of the network traffic, analyzing and simulating for the network dynamic, and improving the predictive ability for system dynamics. In this study, the ensemble of the flexible neural tree (FNT) and system models expressed by the ordinary differential equations (ODEs) is proposed to further improve the accuracy of time series forecasting. Firstly, the additive tree model is introduced to represent more precisely ODEs for the network dynamics. Secondly, the structures and parameters of FNT and the additive tree model are optimized based on the Genetic Programming (GP) and the Particle Swarm Optimization algorithm (PSO). Finally, the expected level of performance is verified by using the proposed method, which provides a reliable forecast model for small-time scale network traffic. Experimental results reveal that the proposed method is able to estimate the small-time scale network traffic measurement data with decent accuracy.

Credit Risk Assessment Based on Flexible Neural Tree Model

In recent years, as China’s credit market continues to expand, a large number of P2P (person-to-person borrow or lend money in Internet Finance) platforms were born and developed. Most of the P2P platforms in China use data mining methods to evaluate the credit risk of loan applicants. Artificial neural network (ANN) is an emerging data mining tool and has good classification ability in many application fields. This paper presents a model of credit risk assessment based on flexible neural tree (FNT), which can reduce the overdue rate and save the analysis time. Overdue and non-overdue sample data are provided by the Jinan Hengxin Micro-Investment Advisory Co., Ltd., and used to build the model. Experiments show that the proposed model is more accurate and has less time cost for the overdue classification of credit risk assessment.

The neoteric feature extraction method of epilepsy EEG based on the vertex strength distribution of weighted complex network

The study of epilepsy detection has great clinical significance. The focus of this study is feature extraction method, which has significant impacts on the performance of epilepsy detection. Recently, the statistic properties of complex network show ability to describe the dynamics of nonlinear time series. In this paper, a feature extraction method of epileptic EEG, based on statistical properties of weighted complex network, is proposed. The weighted network of epileptic EEG is first constructed and the vertex strength distribution of the converted network is studied. Then the weighted mean value of the vertex strength distribution is defined and extracted as the classification feature. Experimental results indicate that the extracted feature can clearly reflect the difference between ictal EEGs and interictal EEGs and the single feature classification based on extracted feature gets higher classification accuracy up to 95.50%.

Local prediction of network traffic measurements data based on relevance vector machine

In the reconstructed phase space, based on the nonlinear time series local prediction method and the relevance vector machine model, the local relevance vector machine prediction method was proposed in this paper, which was applied to predict the small scale traffic measurements data. The experiment results show that the local relevance vector machine prediction method could effectively predict the small scale traffic measurements data, the prediction error mainly concentrated on the vicinity of zero, and the prediction accuracy of the local relevance vector machine regression model was superior to that of the feedforward neural network optimized by PSO.

Seizure detection in clinical EEG based on multi-feature integration and SVM

Recurrence Quantification Analysis (RQA) was a nonlinear analysis method and widely used to analyze EEG signals. In this work, a feature extraction method based on the RQA measures was proposed to detect the epileptic EEG from EEG recordings. To combine the time-frequency characteristic of epileptic EEG, variation coefficient and fluctuation index were used to analyze epileptic EEG. The multi-feature combination of RQA and linear parameters had better performance in analyzing the nonlinear dynamic characteristics and time-frequency characteristic of epileptic EEG. For features selection and improving the classification accuracy, a support vector machine (SVM) classifier was used. The experimental results showed that the proposed method could classify the ictal EEG and interictal EEG with accuracy of 97.98%.