Jianwen Xie

Feature Selection Algorithm Based on Association Rules Mining Method

This paper presents a novel feature selection algorithm based on the technique of mining association rules. The main idea of the proposed algorithm is to find the features that are closely correlative with the class attribute by association rules mining method. Experimental results on several real and artificial data sets demonstrate that the proposed feature selection algorithm is able to obtain a smaller and satisfactory feature subset when compared with other existing feature selection algorithms. It is a new feature selection algorithm with vast of application prospect and research value.

Learning Sparse FRAME Models for Natural Image Patterns

It is well known that natural images admit sparse representations by redundant dictionaries of basis functions such as Gabor-like wavelets. However, it is still an open question as to what the next layer of representational units above the layer of wavelets should be. We address this fundamental question by proposing a sparse FRAME (Filters, Random field, And Maximum Entropy) model for representing natural image patterns. Our sparse FRAME model is an inhomogeneous generalization of the original FRAME model. It is a non-stationary Markov random field model that reproduces the observed statistical properties of filter responses at a subset of selected locations, scales and orientations. Each sparse FRAME model is intended to represent an object pattern and can be considered a deformable template. The sparse FRAME model can be written as a shared sparse coding model, which motivates us to propose a two-stage algorithm for learning the model. The first stage selects the subset of wavelets from the dictionary by a shared matching pursuit algorithm. The second stage then estimates the parameters of the model given the selected wavelets. Our experiments show that the sparse FRAME models are capable of representing a wide variety of object patterns in natural images and that the learned models are useful for object classification.

Vehicle Routing Problem with Time Windows: A Hybrid Particle Swarm Optimization Approach

Vehicle routing problem (VRP) is a well-known combinatorial optimization and nonlinear programming problem seeking to service a number of customers with a fleet of vehicles. This paper proposes a hybrid particle swarm optimization (HPSO) algorithm for VRP. The proposed algorithm utilizes the crossover operation that originally appears in genetic algorithm (GA) to make its manipulation more readily and avoid being trapped in local optimum, and simultaneously for improving the convergence speed of the algorithm, level set theory is also added to it. We employ the HPSO algorithm to an example of VRP, and compare its result with those generated by PSO, GA, and parallel PSO algorithms. The experimental comparison results indicate that the performance of HPSO algorithm is superior to others, and it will become an effective approach for solving discrete combinatory problems.

A particle swarm optimization algorithm with crossover for vehicle routing problem with time windows

The vehicle routing problem (VRP) is a very important combinatorial optimization and nonlinear programming problem in the fields of transportation, distribution and logistics. In this paper, a particle swarm optimization (PSO) algorithm with crossover for VRP is proposed. The PSO algorithm combined with the crossover operation of genetic algorithm (GA) can avoid being trapped in local optimum due to using probability searching. We apply the proposed algorithm to an example of VRP, and compare its result with those generated by PSO, GA, and parallel PSO algorithms. The experimental comparison result demonstrates that the performance of PSO algorithm with crossover is competitive with others and will be an effective method for solving discrete combinatory problems.

A Particle Swarm Optimization Algorithm Based on Diffusion-Repulsion and Application to Portfolio Selection

Inspired by the diffusion movement phenomenon of the molecules, this paper presents a diffusion-repulsion particle swarm optimization (DRPSO). The proposed new algorithm (DRPSO) includes attraction and repulsion (or migration) phases. Once the diversity of population becomes too low, the individuals will be dispersed and carry out diffusion movement, while if the diversity of population becomes too high, the individuals have to be congregated, which is accomplished by diversity control method. Comparisons with standard SPSO and other algorithms on a portfolio problem indicate that DRPSO not only prevents premature convergence to a high degree, but also keeps a more rapid convergence rate than SPSO.

Decoding the encoding of functional brain networks: An fMRI classification comparison of non-negative matrix factorization (NMF), independent component analysis (ICA), and sparse coding algorithms

BACKGROUND Brain networks in fMRI are typically identified using spatial independent component analysis (ICA), yet other mathematical constraints provide alternate biologically-plausible frameworks for generating brain networks. Non-negative matrix factorization (NMF) would suppress negative BOLD signal by enforcing positivity. Spatial sparse coding algorithms (L1 Regularized Learning and K-SVD) would impose local specialization and a discouragement of multitasking, where the total observed activity in a single voxel originates from a restricted number of possible brain networks. NEW METHOD The assumptions of independence, positivity, and sparsity to encode task-related brain networks are compared; the resulting brain networks within scan for different constraints are used as basis functions to encode observed functional activity. These encodings are then decoded using machine learning, by using the time series weights to predict within scan whether a subject is viewing a video, listening to an audio cue, or at rest, in 304 fMRI scans from 51 subjects. RESULTS AND COMPARISON WITH EXISTING METHOD The sparse coding algorithm of L1 Regularized Learning outperformed 4 variations of ICA (p<0.001) for predicting the task being performed within each scan using artifact-cleaned components. The NMF algorithms, which suppressed negative BOLD signal, had the poorest accuracy compared to the ICA and sparse coding algorithms. Holding constant the effect of the extraction algorithm, encodings using sparser spatial networks (containing more zero-valued voxels) had higher classification accuracy (p<0.001). Lower classification accuracy occurred when the extracted spatial maps contained more CSF regions (p<0.001). CONCLUSION The success of sparse coding algorithms suggests that algorithms which enforce sparsity, discourage multitasking, and promote local specialization may capture better the underlying source processes than those which allow inexhaustible local processes such as ICA. Negative BOLD signal may capture task-related activations.

An image template matching method using particle swarm optimization

At present, most of the image template matching algorithms involve large computational complexity, and can hardly be used in practical projects. This paper proposes that particle swarm optimization algorithm (PSO) be used in image template matching problems (ITMP). Template matching in fact is a matter of seeking optimization. The cross-correlation function of template and sub image is set as the objective function, and then a fast template matching algorithm can reached based on particle swarm optimization algorithm. Experiment results prove both the computational accuracy and efficiency of this algorithm.

A Particle Swarm Optimization Algorithm with Crossover for Resource Constrained Project Scheduling Problem

Resource and project optimization scheduling has become the key of the success of researching project in the enterprises. In order to solve the mass resource constrained project scheduling problem, in this paper, an improved particle swarm algorithm (PSO) called particle swarm algorithm with crossover (CPSO) was presented. This improved algorithm is based on PSO and genetic algorithm (GA). Through comparing with SPSO and GA on RCPSP, it is indicated that CPSO not only avoids premature convergence to a high degree, but also keeps a faster convergence rate than SPSO and GA.

Synthesizing Dynamic Patterns by Spatial-Temporal Generative ConvNet

Video sequences contain rich dynamic patterns, such as dynamic texture patterns that exhibit stationarity in the temporal domain, and action patterns that are non-stationary in either spatial or temporal domain. We show that a spatial-temporal generative ConvNet can be used to model and synthesize dynamic patterns. The model defines a probability distribution on the video sequence, and the log probability is defined by a spatial-temporal ConvNet that consists of multiple layers of spatial-temporal filters to capture spatial-temporal patterns of different scales. The model can be learned from the training video sequences by an analysis by synthesis learning algorithm that iterates the following two steps. Step 1 synthesizes video sequences from the currently learned model. Step 2 then updates the model parameters based on the difference between the synthesized video sequences and the observed training sequences. We show that the learning algorithm can synthesize realistic dynamic patterns.

An Improved Grey-Markov Chain Method with an Application to Predict the Number of Chinese International Airlines

This paper proposes an improved Grey-Markov forecasting dynamic method based on unbiased grey system theory and fuzzy classification. The new forecasting method is named unbiased Grey-fuzzy-Markov Chain method, which can take advantage of the prediction power of conventional Grey-Markov forecasting method and at the same time eliminate grey bias and improve anti-jamming performance. As an example, we use the statistical data of the number of Chinese international airlines from 1987 to 2006 for a validation of the feasibility and practicability of the improved Grey-Markov forecasting model.