检索结果-内蒙古大学图书馆

Detect epistasis based on optimization of artificial neural network architecture and parameters

ICIC Express Letters 2015年第5期9卷 1499-1506页

作者： Zhou, Zhihui Liu, Guixia Su, Lingta Tian, Yuan College of Computer Science and Technology China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University No. 2699 Qianjin Avenue Changchun China

Epistasis or the interaction of single nucleotide polymorphisms (SNPs) at different loci plays a significant role on the mechanisms and pathogenesis of many common complex, multifactorial diseases and may be responsible for the “missing” heredity in genetic epidemiology. A certain complex network of epistatic interactions may contribute to the disease susceptibility. Epistasis detection is a computational and statistical challenge for the absence of prior knowledge of biological data, intensive computational burden and inappropriate statistical tests. Here we investigate the application of a standard artificial neural network (ANN) to detect epistasis for genome-wide association studies (GWAS) for the ability of ANN to model nonlinear data. The pattern-recognition strategy of ANN is utilized in the hope that genotypes of SNP markers will successfully distinguish between cases and controls in a way leading to the ANN being able to classify the SNPs correctly. Using simulated data under a wide range of settings, we show that the proposed ANN can well model gene-gene interactions with good predictive ability. The performance of ANN shows considerable promise and further improvements could usefully be designed towards reducing the computing speed and optimizing implementation of such networks. © 2015, ICIC International.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Document compression scheme based on integer data

引用

Journal of computational Information Systems 2014年第12期10卷 5203-5210页

作者： Liu, Yang Te, Rigen Li, Xiongfei Key Laboratory of Symbolic Computation and Knowledge Engineering for Ministry of Education Jilin University Changchun 130012 China

Data compression does not only save space for data storage, but also improve its safety and efficiency during data transport. As any data can be saved in the form of an integer directly or indirectly, it is a meaningful work for us to extend integer data compression technique to the application of any arbitrary data compression. In this paper we propose a new compression algorithm (CNA), which does not need additional data support and data compression and reduction are lossless, for integer data based on Godel numbers. In the meantime, we will test the approach and show the uniqueness of the compression algorithm result and the correctness of the decompression data. By contrastive experimentation, the CNA compression algorithm proves good result for any documents, especially for integer documents. 1553-9105/Copyright © 2014 Binary Information Press.

关键词： Data compression

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Multi-path Routing for VANET

A Dynamic Multi-path Routing for VANET

引用

2015 International Conference on Automation,Mechanical and Electrical engineering(AMEE 2015)

作者： Z.Y.Yuan D.Wei J.Q.Zhu Y.J.Hou M.Li T.Sun College of Computer Science and Technology Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun Regional Road Transportation Administrative Bureau

Compared with traditional VANET(Vehicular Ad-Hoc Networks) routing techniques, geographic routing has been proven to be more fittable for highly mobile scenes. Traditional routings use greedy modes or fixed forwarding paths to sent packets. But, the dynamic features of VANET such as fast changed topology, vehicles density and radio obstacles, could cause local maximum and sparse connectivity. Due to the characteristics of wireless channel, while there are too many packets transmit through a path, the delay and the number of packets loss will both increase clearly. We propose DMPR, a Dynamic Multipath Routing, combined with node location and digital map. The proposed DMPR detects transmission delay of different paths every once in a period of time, and dynamically determine the transmission ratio of each path. We execute NS2 simulation to exhibit that DMPR routing protocol significantly outperform three well known VANET ones in terms of the average packet delivery ratio and end-to-end delay.

关键词： VANET dynamic multi-path routing digital map transmission delay detection

来源：评论

学校读者我要写书评

暂无评论

An efficient and effective similarity measure to enable data mining of fabric image of Xinjiang folk art

引用

Journal of Information and computational Science 2015年第5期12卷 1695-1702页

作者： Te, Rigen Li, Xiongfei Zhang, Yueying Key Laboratory of Symbol Computation and Knowledge Engineering Jilin University Changchun China College of Materials Science and Engineering Jilin University Changchun China

Fabric pattern is a folk art of artificial fabrics pattern design which includes apparels (known as clothes and hats), and crafts (known as carpets and tapestries). Although image processing, information retrieval and data mining have made a large impact on many human endeavors, they never had an impact on the study of the fabric image. In this paper we will present the reasons for this, and introduce a novel similarity measure and algorithm (Pattern Similarity Measuring Algorithm (PSMA)) which allows efficient data mining for large collections of fabric image. PSMA, the effectiveness of which is proved be experiment test, is an image similarity measurement algorithm that achieves the machine-learning based theory via the gradient descent algorithm with multidimensional linear regression. Copyright © 2015 Binary Information Press.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Bayesian inference for epistasis association mapping in case-control studies

引用

ICIC Express Letters 2015年第1期9卷 261-268页

作者： Zhou, Zhihui Liu, Guixia Su, Lingtao Tian, Yuan College of Computer Science and Technology Jilin University No. 2699 Qianjin Avenue Changchun China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University No. 2699 Qianjin Avenue Changchun China

A genome-wide association study (GWAS) typically involves detecting epistatic interactions of multiple genetic variants on the susceptibility of complex human diseases. As the most abundant source of genetic variation in human genome, the number of single nucleotide polymorphisms (SNPs) reaches millions in public datasets and SNP data collected from thousands of individuals in case-control studies may shed light on our understanding of epistatic interactions. Various methods have been proposed in previous literature for identifying genetic interactions, but they are infeasible for GWAS as biology data is too large, we here propose a method ‘epistasis group based on Bayesian inference’ (EGBI). EGBI applies a Bayesian marker partition model to investigate observed case-control data and computes the posterior distribution of each epistasis group that is associated with the disease via Markov chain Monte Carlo (MCMC). When applied to both simulated data and WTCCC type 1 diabetes data, EGBI successfully identified many known susceptible genes including CTLA4 and MHC and performed more powerfully than its competitors.

关键词： Genes

来源：评论

学校读者我要写书评

暂无评论

Using a random forest method for epistasis detection in genome-wide association studies

引用

Journal of computational Information Systems 2014年第18期10卷 8043-8050页

作者： Zhou, Zhihui Liu, Guixia Su, Lingtao Han, Liang Yan, Lun College of Computer Science and Technology and Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun China

Detecting susceptibility genes and gene-gene interactions (epistasis) is an important issue in genetic association analysis and genetic epidemiology. Due to the huge number of single nucleotide polymorphisms (SNPs) and inappropriate statistical tests, epistasis detection is a computational and statistical challenge and becomes a "needle-in-a-haystack" problem. Also some epistasis detection algorithms proposed in lots of literature have demonstrated their successes for small scale data, while most of them cannot be directly applied into genome-wide association studies (GWAS) and the pathogenesis of many common complex human diseases is mysterious. Here we adopted a random forest method incorporating information theory and a powerful statistical test, B-stat to detect epistasis. We conducted sufficient artificial experiments on a wide range of simulated datasets and compared performance of our random forest method with its two competitors, COE and BEAM. Experimental results demonstrated that this method is quite available and time efficiency for the haystack problem. We also presented the results of the application of the method to the WTCCC type 1 diabetes dataset. We reported some previously well known genes as well as some significant SNP interactions.

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

A survey on virtual network embedding in cloud computing centers

引用

Open Automation and Control Systems Journal 2014年第1期6卷 414-425页

作者： Wei, Xiaohui Hu, Shoufeng Li, Hongliang Yang, Fan Jin, Yue College of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education China

Bridging virtualized environments with physical environments, virtual network plays an important role in Cloud Computing infrastructures. How to allocate physical resources for virtual nodes/links to construct virtual network is known as Virtual Network Embedding (VNE) problem. It is a crucial issue that draws wide attention. This paper surveys existing VNE methods and algorithms and how they work under different application scenarios. First, a survey of different VNE objective metrics is presented;then several VNE algorithms are presented and categorized;finally, the performance of different VNE solutions are compared and discussed. © 2010, Bentham Science Publishers B.V. All rights reserved.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Intelligent Model Learning Based on Variance for Bayesian Reinforcement Learning

Intelligent Model Learning Based on Variance for Bayesian Re...

引用

International Conference on Tools with Artificial Intelligence

作者： Shuhua You Quan Liu Zongzhang Zhang Hui Wang Xiaofang Zhang School of Computer Science and Technology Soochow University Suzhou China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education

ISBN: (纸本)9781509001644

We consider a modular method to reinforcement learning that represents uncertainty of model parameters by maintaining probability distributions over them. The algorithm we call MBDP (model-based Bayesian dynamic programming) can be decomposed into two parallel types of inference: model learning and policy learning. During learning a model, we update posterior distributions of a model over observations after taking an action in each state. During learning a policy, we solve MDPs by dynamic programming with greedy approximation to make an agent choose behaviors which maximize return under the estimated model. Furthermore, we propose a principled method which utilizes the variance of Dirichlet distributions for determining when to learn and relearn the model. We demonstrate that MBDP can find near optimal policies with high probability by sufficient model learning and experimental results show that MBDP performs better compared with current state-of-the-art methods in reinforcement learning.

关键词： reinforcement learning Bayesian dynamic programming model learning policy learning Dirichlet distributions

来源：评论

学校读者我要写书评

暂无评论

Methods for gene selection in microarrays based on data clustering on local SVM-RFE

引用

Journal of computational Information Systems 2015年第16期11卷 5825-5832页

作者： Cao, Zhongbo Du, Wei Sun, Ying Liang, Yanchun Key Laboratory of Symbolic Computation and Knowledge Engineering of the Ministry of Education College of Computer Science and Technology Jilin University Changchun China Department of Computer Science and Technology Zhuhai College of Jilin University Zhuhai China School of Management Science and Information Engineering Jilin University of Finance and Economics Changchun China

Most of the microarray expression data have tens of thousands of genes but very small number of samples. Feature selection has been widely used to extract the subset of informative genes. Though many feature selection methods have been proposed, the Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithm is one of the best for microarray expression data analysis. However, SVM-RFE is not a stable and robust algorithm in some cases. The purpose of this article is to design a novel gene selection method for microarray expression data analysis, which is more effective, flexible and robust than normal SVM-RFE. In this article, the Local SVM-RFE (L-SR), Cluster SVM-RFE (C-SR) and Cluster Local SVM-RFE (CL-SR) methods for feature selection are proposed which eliminate irrelevant genes effectively and automatically. Experimental results obtained by using six cancer related datasets on different classifiers demonstrate that the proposed methods in this article are superior to normal SVM-RFE. Moreover, compared with the mRMR, GA-KNN and SVM-RCE, the proposed algorithm shows better performance in best and average accuracies comparison. © 2015 by Binary Information Press

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Ant colony algorithm based on local mutation

引用

Journal of Chemical and Pharmaceutical Research 2014年第8期6卷 508-516页

作者： Shen, Xuan-Jing Shen, Xuan-Jing College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun China

Ant colony algorithm is a more effective way of solving traveling salesman problem (TSP). Ant colony algorithm adopts the distributed parallel computing mechanism;and it is easy to combine with other methods. Furthermore, it has the advantages of strong robustness. But the search time is long, and it often falls into local optimal solution. In order to speed up convergence rate and ensure the global search ability, use adaptive adjustment mechanism to adjust pheromone evaporation factor, and mutate solutions according to the characteristics of ant colony searching for path in each iteration. An improved ant colony algorithm is proposed in this paper. Experiments show that algorithm has a more remarkable quality of precision and convergent velocity than T. A. S. Masutti and G. F. Dong’s, the improved Percentage deviations of average to the BKS is less than 1% equally, percentage deviations of best to the BKS is less than 0.05% equally. © 2014, Journal of Chemical and Pharmaceutical Research. All rights reserved.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：