Huijuan Lu

A hybrid feature selection algorithm for gene expression data classification

In the DNA microarray research field, the increasing sample size and feature dimension of the gene expression data prompt the development of an efficient and robust feature selection algorithm for gene expression data classification. In this study, we propose a hybrid feature selection algorithm that combines the mutual information maximization (MIM) and the adaptive genetic algorithm (AGA). Experimental results show that the proposing MIMAGA-Selection method significantly reduces the dimension of gene expression data and removes the redundancies for classification. The reduced gene expression dataset provides highest classification accuracy compared to conventional feature selection algorithms. We also apply four different classifiers to the reduced dataset to demonstrate the robustness of the proposed MIMAGA-Selection algorithm.

A cost-sensitive rotation forest algorithm for gene expression data classification

Existing works show that the rotation forest algorithm has competitive performance in terms of classification accuracy for gene expression data. However, most existing works only focus on the classification accuracy and neglect the classification costs. In this study, we propose a cost-sensitive rotation forest algorithm for gene expression data classification. Three classification costs, namely misclassification cost, test cost and rejection cost, are embedded into the rotation forest algorithm. This extension of the rotation forest algorithm is named as cost-sensitive rotation forest algorithm. Experimental results show that the cost-sensitive rotation forest algorithms effectively reduce the classification cost and make the classification result more reliable.

Fast and Accurate Classification of Time Series Data Using Extended ELM: Application in Fault Diagnosis of Air Handling Units

The extreme learning machine (ELM) is famous for its single hidden-layer feed-forward neural network which results in much faster learning speed comparing with traditional machine learning techniques. Moreover, extensions of ELM achieve stable classification performances for imbalanced data. In this paper, we introduce a hybrid method combining the extended Kalman filter (EKF) with cost-sensitive dissimilar ELM (CS-D-ELM). The raw data are preprocessed by EKF to produce inputs for the CS-D-ELM classifier. Experimental results show that the proposed method is more suitable for real-time fault diagnosis of air handling units than traditional approaches.

Learning Misclassification Costs for Imbalanced Datasets, Application in Gene Expression Data Classification

Cost-sensitive algorithms have been widely used to solve imbalanced classification problem. However, the misclassification costs are usually determined empirically, leading to uncertain performance. Hence an effective method is desired to automatically calculate the optimal cost weights. Targeting at the highest weighted classification accuracy (WCA), we propose two approaches to search for the optimal cost weights, including grid searching and function fitting. In experiments, we classify imbalanced gene expression data using extreme learning machine to test the cost weights obtained by the two approaches. Comprehensive experimental results show that the function fitting is more efficient which can well find the optimal cost weights with acceptable WCA.

Kernel principal component analysis combining rotation forest method for linearly inseparable data

Abstract Rotation forest (RoF) is an ensemble classifier combining linear analysis theories and decision tree algorithms. In recent existing works, RoF was widely applied to various fields with outstanding performance compared to traditional machine learning techniques, given that a reasonable number of base classifiers is provided. However, the conventional RoF algorithm suffers from classifying linearly inseparable datasets. In this study, a hybrid algorithm integrating kernel principal component analysis (KPCA) and the conventional RoF algorithm is proposed to overcome the classification difficulty for linearly inseparable datasets. The radial basis function (RBF) is selected as the kernel for the KPCA method to establish the nonlinear mapping for linearly inseparable data. Moreover, we evaluate various kernel parameters for better performance. Experimental results show that our algorithm improves the performance of RoF with linearly inseparable datasets, and therefore provides higher classification accuracy rates compared with other ensemble machine learning methods.

Classifying Non-linear Gene Expression Data Using a Novel Hybrid Rotation Forest Method

Rotation forest (RoF) is an ensemble classifier based on the combination of linear analysis theories and decision tree algorithms. In existing works, the RoF has demonstrated high classification accuracy and good performance with a reasonable number of base classifiers. However, the classification accuracy drops drastically for linearly inseparable datasets. This paper presents a hybrid algorithm integrating kernel principal component analysis and RoF algorithm (KPCA-RoF) to solve the classification problem in linearly inseparable cases. We choose the radial basis function (RBF) kernel for the PCA algorithm to establish the nonlinear mapping and segmentation for gene data. Moreover, we focus on the determination of suitable parameters in the kernel functions for better performance. Experimental results show that our algorithm solves linearly inseparable problem and improves the classification accuracy.

A Method of Determining the Cost Weight with High-Dimensional Fitting

Cost-sensitive algorithms are widely used to solve unbalanced classification problems. However, cost-weight parameters are usually set according to experiences. The optimal parameters remain unsure, which affect the final classification performance. This work utilizes balance accuracy as the evaluation standard, obtains the classification accuracy under different weights settings by adaptive algorithm, and eventually obtains the optimal cost-weight function with highest classification accuracy through 3d fitting. In the experiment, we classify gene expression data with cost-weight parameters obtained by cost-weight function; and the results show that the proposed algorithm is widely applicable to various imbalanced datasets.

ChildGuard: A Child-Safety Monitoring System

With the rapid development of urbanization and industrialization in China, more and more children are studying and living in cities, which presents some safety challenges. To help guardians better monitor their children, the authors present ChildGuard, a child safety system based on mobile devices. ChildGuard provides an in-path safety function that monitors the real-time movement of children walking on the road. It also provides a region safety function that sets designated areas in which children can play. Children can be warned about potential risks, and their guardians can be informed of location or activity abnormities. Experiments show that ChildGuard has higher positioning accuracy and better real-time communication than similar systems. This article is part of a special issue on cybersecurity.

An Analogous Teaching Method for Computer Organization Course Design

Computer Organization Course Design is a major course in computer and related specialties. This paper presents an analogous teaching method which combining simulators with FPGA (Field-Programmable Gate Array) platforms. The analogous teaching method includes the Global-Local and Local-Global teaching method and the from CISC (Complex Instruction Set Computer) to RISC (Reduced Instruction Set Computer) teaching method, the organization method based on groups, the realization method combining after-class programming and in-class verification, and the systematic evaluation method. By comparing the difference and analogies of the two platforms, the analogous teaching method enhances the understanding of students on the difference implementation and working mechanisms of computers, improves the ability to combine theories with practices, and develops the ability to model, design, and implement components and systems of computers by using EDA (Electronic Design Automation) tools.

Terminal neural computing

To solve the time-varying Sylvester equations, a special kind of terminal neural networks (TNN) and its accelerated form are presented, which show better convergent behaviors of asymptotic ones. The terminal attraction of the matrix differential equations is analyzed, and the results show that the method can assure the networks of converging to zero during a limited period. The terminal neural networks can also be used to account for the time-varying matrix inversion as well as the trajectory planning of redundant manipulators. The typical example for a planar is the manipulator in which the end-effector appeared as a closed path, and the joint variables can return to the initial values, making the motion repeatable. The simulation results certify for the validity and superiority of the terminal neural method.