Time-series forecasting is an important research and application area. Much effort has been devoted over the past several decades to develop and improve the time-series forecasting models. This paper introduces a new time-series forecasting model based on the flexible neural tree (FNT). The FNT model is generated initially as a flexible multi-layer feed-forward neural network and evolved using an evolutionary procedure. Very often it is a difficult task to select the proper input variables or time-lags for constructing a time-series model. Our research demonstrates that the FNT model is capable of handing the task automatically. The performance and effectiveness of the proposed method are evaluated using time series prediction problems and compared with those of related methods.

Time-series forecasting using flexible neural tree model

Reviewing ensemble classification methods in breast cancer.

The Elman neural network has good dynamic properties and strong global stability, being most widely used to deal with nonlinear, dynamic, and complex data. However, as an optimization of the backpropagation (BP) neural network, the Elman model inevitably inherits some of its inherent deficiencies, influencing the recognition precision and operating efficiency. Many improvements have been proposed to resolve these problems, but it has proved difficult to balance the many relevant features such as storage space, algorithm efficiency, recognition precision, etc. Also, it is difficult to obtain a permanent solution from a temporary solution simultaneously. To address this, a genetic algorithm (GA) can be introduced into the Elman algorithm to optimize the connection weights and thresholds, which can prevent the neural network from becoming trapped in local minima and improve the training speed and success rate. The structure of the hidden layer can also be optimized using the GA, which can solve the difficult problem of determining the number of neurons. Most previous studies on such evolutionary Elman algorithms optimized the connection weights or network structure individually, which represents a slight deficiency. We propose herein a novel optimized GA–Elman neural network algorithm where the connection weights are real-encoded, while the neurons of the hidden layer also adopt real-coding but with the addition of binary control genes. In this new algorithm, the connection weights and the number of hidden neurons are optimized using hybrid encoding and evolution simultaneously, greatly improving the performance of the resulting novel GA–Elman algorithm. The results of three experiments show that this new GA–Elman model is superior to the traditional model in terms of all calculated indexes.

A novel optimized GA–Elman neural network algorithm

This paper provides a comprehensive review regarding the applications of particle swarm optimization (PSO) in the railway domain. One hundred and thirty nine (139) publications in the railway domain are listed and summarized. The review indicates that PSO has seen more and more applications in the railway domain in recent years; scheduling, active controls, and network layout planning represent the three largest application areas. PSO variants such as genetic PSO, chaotic PSO, and quantum-behaved PSO are also used in the railway domain. The inertial weight has been widely accepted and used in railway applications, while the contraction coefficient and variable velocity limit have seen fewer applications. Optimization of vehicle mechanical systems dynamics has been identified as an area that has the potential for more applications. From this paper, researchers from other areas of the railway domain can identify many other potential applications. Parallel PSO was not found in previous railway applic...

Applications of particle swarm optimization in the railway domain

On the one hand, swarm intelligence SI is an emerging field of artificial intelligence that takes inspiration in the collective and social behavior of different groups of simple agents. On the other hand, the automatic evolution of programs is an active research area that has attracted a lot of interest and has been mostly promoted by the genetic programming paradigm. The main objective is to find computer programs from a high-level problem statement of what needs to be done, without needing to know the structure of the solution beforehand. This paper looks at the intersection between SI and automatic programming, providing a survey on the state-of-the-art of the automatic programming algorithms that use an SI metaheuristic as the search technique. The expression of swarm programming SP has been coined to cover swarm-based automatic programming proposals, since they have been published to date in a disorganized manner. Open issues for future research are listed. Although it is a very recent area, we hope that this work will stimulate the interest of the research community in the development of new SP metaheuristics, algorithms, and applications. WIREs Data Mining Knowl Discov 2014, 4:445-469. doi: 10.1002/widm.1138

Swarm-based metaheuristics in automatic programming: a survey

Neural tree model has been successfully applied to solving a variety of interesting problems. In most previous studies, optimization of the neural tree model was divided into two steps: first structure optimization, then parameter optimization. One major problem in the evolution of structure without parameter information was noisy fitness evaluation. In this paper, an improved breeder genetic programming algorithm is proposed to the synthesis of neural tree model. The effectiveness and performance of the method are evaluated on time series prediction problems and compared with those of related methods. Simulation results show that the proposed algorithm is a potential method with better performance and effectiveness.

Synthesis of neural tree models by improved breeder genetic programming

The railway passenger traffic volume(RPTV) forecast can offer scientific basis for the establishment of policy and making of transportation development plan. This paper applies the neural tree model for predicting the railway passenger traffic volume. The optimal structure is developed using the Improved Probabilistic Incremental Program Evolution(IPIPE) and the free parameters encoded in the optimal tree are optimized by the Particle Swarm Optimization(PSO) algorithm, and an improved sigmoid function is applied as the neural activation function, a new fitness function combines error and Occam's razor is used for for balancing of accuracy and parsimony of evolved structures. Based on the RPTV from 1985 to 2007 of China, the performance and efficiency of the applied model are evaluated and compared with the multi-layer feed-forward network(MLFN) and support vector machine(SVM).

Prediction of Railway Passenger Traffic Volume Based on Neural Tree Model

By changing the linear encoding and redefining the evolving rules, particle swarm algorithm is introduced into genetic programming and an hybrid genetic programming with particle swarm optimization (HGPPSO) is proposed. The performance of the proposed algorithm is tested on tow symbolic regression problem in genetic programming and the simulation results show that HGPPSO is better than genetic programming in both convergence times and average convergence generations and is a promising hybrid genetic programming algorithm.

A Hybrid Genetic Programming with Particle Swarm Optimization

A neural tree network ensemble model is proposed for classification which is an important research field in data mining and machine learning. Firstly, establishes each single neural tree network by using an improved hybrid breeder genetic programming, and then more neural tree networks are combined to form the final classification model by the idea of ensemble learning. Simulation results on two disease classification problems show that this model is effective for the classification, and has better performance in classification precision, feature selection and structure simplification, especially for classification with multi-class attributes.

A Neural Tree Network Ensemble Mode for Disease Classification

Since the subprime crisis, the variance of housing price is receiving increasing attention especially because of its complexity and practical applications. This paper applies the flexible neural tree model for forecasting the housing price index (HPI). The optimal structure is developed using the Modified Breeder Genetic Programming (MBGP) and the free parameters encoded in the optimal tree are optimized by the Particle Swarm Optimization (PSO), and a new fitness function based on error and Occam's razor is used for for balancing of accuracy and parsimony of evolved structures. Based on the HPI of Shandong province, the performance and efficiency of the applied model are evaluated and compared with the classical multi-layer feed-forward network (MLFN) and support vector machine (SVM) models.

Feng Qi

Papers

Synthesis of neural tree models by improved breeder genetic programming

Prediction of Railway Passenger Traffic Volume Based on Neural Tree Model

A Hybrid Genetic Programming with Particle Swarm Optimization

A Neural Tree Network Ensemble Mode for Disease Classification

Housing price index forecasting using neural tree model