Yaou Zhao

A novel ensemble of classifiers for microarray data classification

Micorarray data are often extremely asymmetric in dimensionality, such as thousands or even tens of thousands of genes and a few hundreds of samples. Such extreme asymmetry between the dimensionality of genes and samples presents several challenges to conventional clustering and classification methods. In this paper, a novel ensemble method is proposed. Firstly, in order to extract useful features and reduce dimensionality, different feature selection methods such as correlation analysis, Fisher-ratio is used to form different feature subsets. Then a pool of candidate base classifiers is generated to learn the subsets which are re-sampling from the different feature subsets with PSO (Particle Swarm Optimization) algorithm. At last, appropriate classifiers are selected to construct the classification committee using EDAs (Estimation of Distribution Algorithms). Experiments show that the proposed method produces the best recognition rates on four benchmark databases.

Stock Index Modeling using EDA based Local Linear Wavelet Neural Network

The use of intelligent systems for stock market predictions has been widely established. In this paper, we investigate how the seemingly chaotic behavior of stock markets could be well represented using local linear wavelet neural network (LLWNN) technique. To this end, we considered the Nasdaq-100 index of Nasdaq Stock MarketSM and the S&P CNX NIFTY stock index. We analyzed 7-year Nasdaq-100 main index values and 4-year NIFTY index values. This paper investigates the development of novel reliable and efficient techniques to model the seemingly chaotic behavior of stock markets. The LLWNN are optimized by using estimation of distribution algorithm (EDA). This paper investigates whether the proposed method can provide the required level of performance, which is sufficiently good and robust so as to provide a reliable forecast model for stock market indices. Experiment results shown that the model considered could represent the stock indices behavior very accurately

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.

A novel method for prediction of protein interaction sites based on integrated RBF neural networks

Protein interactions are very important for control life activities. If we want to study the principle of protein interactions, we have to find the seats of a protein which are involved in the interactions called interaction sites firstly. In this paper, a novel method based on an integrated RBF neural networks is proposed for prediction of protein interaction sites. At first, a number of features were extracted, i.e., sequence profiles, entropy, relative entropy, conservation weight, accessible surface area and sequence variability. Then 6 sliding windows about these features were made, and they contained 1, 3, 5, 7, 9 and 11 amino acid residues respectively. These sliding windows were put into the input layers of six radial basis functional neural networks that were optimized by Particle Swarm Optimization. Thus, six group results were obtained. Finally, these six group results were integrated by decision fusion (DF) and Genetic Algorithm based Selective Ensemble (GASEN). The experimental results show that the proposed method performs better than the other related methods such as neural networks and support vector machine.

A Novel Ensemble Approach for Cancer Data Classification

Micorarray data are often extremely asymmetric in dimensionality, such as thousands or even tens of thousands of genes and a few hundreds of samples. Such extreme asymmetry between the dimensionality of genes and samples presents several challenges to conventional clustering and classification methods. In this paper, a novel ensemble method based on correlation analysis is proposed. Firstly, in order to extract useful features and reduce dimensionality, different feature selection methods based on correlation analysis are used to form different feature subsets. Then a pool of candidate base classifiers is generated to learn the subsets which are re-sampling from the different feature subsets. At last, appropriate classifiers are selected to construct the classification committee using EDA (Estimation of Distribution Algorithms) algorithm. Experiments show that the proposed method produces the best recognition rates on two benchmark databases.

Face recognition using DCT and hierarchical RBF model

This paper proposes a new face recognition approach by using the Discrete Cosine Transform (DCT) and Hierarchical Radial Basis Function Network (HRBF) classification model. The DCT is employed to extract the input features to build a face recognition system, and the HRBF is used to identify the faces. Based on the pre-defined instruction/operator sets, a HRBF model can be created and evolved. This framework allows input features selection. The HRBF structure is developed using Extended Compact Genetic Programming (ECGP) and the parameters are optimized by Differential Evolution (DE). Empirical results indicate that the proposed framework is efficient for face recognition.

Multi-layer Ensemble Classifiers on Protein Secondary Structure Prediction

Protein secondary structure prediction is a bridge between amino acid sequence and tertiary prediction. Many methods have been used to improve the prediction accuracy and have a great development. Here a new method to predict the secondary structure is proposed by using classified database. The structure is similar with PSI-PRED method. But in the middle of the two neural network (NN) a neural ensemble is used to divided the initial database into two databases. Then the two databases are input into the next NN separately in order to predict the possible secondary structure. Experiments show that the proposed method achieves an average Q 3 score of 82% in CB396 database and 86% in CB513 database which produces the best prediction 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.

Feature Combination Methods for Prediction of Subcellular Locations of Proteins with Both Single and Multiple Sites

Effective feature extraction methods play very important role for prediction of multisite protein subcellular locations. With the progress of many proteome projects, more and more proteins are annotated with more than one subcellular location. However, compared with the problems of single-site protein, the problems of multiplex protein subcellular localizations are far more difficult and complicated to deal with. To improve the multisite prediction quality, it is necessary to incorporate different feature extraction methods. In this paper, a version of feature combination method which is to make use of the 20 dimensions of entropy density instead of the former 20 dimensions of amphiphilic pseudo amino acid composition (AmPseAAC), is used in two different datasets. It is different from the way of simple dimensions additive feature fusion. On base of this novel feature combination method, we adopt the multi-label k-nearest neighbors (ML-KNN) algorithm and setting different weights into different attributes’ ML-KNN, which is called wML-KNN, to predict multiplex protein subcellular locations. The best overall accuracy rate on dataset S1 from the predictor of Virus-mPLoc is 61.11 % and 82.03 % on dataset S2 from Gpos-mPLoc, respectively.

Stochastic system identification by evolutionary algorithms

For system identification, the ordinary differential equations (ODEs) model is popular for its accuracy and effectiveness. Consequently, the ODEs model is extended to the stochastic differential equations (SDEs) model to tackle the stochastic case intuitively. But the existence of stochastic integral is a rigid barrier. We simply transform the SDEs to their corresponding stochastic difference equations (SDCEs) to eliminate stochastic integrals and propose an easy but effective solution to stochastic system identification. In this solution, the maximum likelihood estimation can be applied and the evolutionary algorithms are used to determine structures and parameters of the unknown system.