一种结合上下文语义的短文本聚类算法

短文本因具有特征信息不足且高维稀疏等特点,使得传统文本聚类算法应用于短文本聚类任务时性能有限。针对上述情况,提出一种结合上下文语义的短文本聚类算法。首先借鉴社会网络分析领域的中心性和权威性思想设计了一种结合上下文语义的特征词权重计算方法,在此基础上构建词条-文本矩阵;然后对该矩阵进行奇异值分解,进一步将原始特征词空间映射到低维的潜在语义空间;最后通过改进的K-means聚类算法在低维潜在语义空间完成短文本聚类。实验结果表明,与传统的基于词频及逆向文档频权重的文本聚类算法相比,该算法能有效改善短文本特征不足及高维稀疏性,提高了短文的本聚类效果。     

Higher order dispersive and dissipative hybrid method for the numerical solution of oscillatory problems

In this paper, an improved explicit two-step hybrid method with fifth algebraic order is derived. The new method possesses dispersion of order 10 and dissipation of order seven, which is first of its kind in the literature. Numerical experiment reveals the superiority of the new method for solving oscillatory or periodic problems over several methods of the same algebraic order.     

Design of an adaptive fuzzy sliding mode control for uncertain discrete-time nonlinear systems based on noisy measurements

This paper presents the design of an adaptive fuzzy sliding mode control (AFSMC) for uncertain discrete-time nonlinear dynamic systems. The dynamic systems are described by a discrete-time state equation with nonlinear uncertainties, and the uncertainties include the modelling errors and the external disturbances to be unknown but nonlinear with the bounded properties. The states are measured by the restriction of measurement sensors and the contamination with independent measurement noises. The nonlinear uncertainties are approximated by using the fuzzy IF-THEN rules based on the universal approximation theorem, and the approximation error is compensated by adding an adaptive complementary term to the proposed AFSMC. The fuzzy inference approach based on the extended single input rule modules is proposed to reduce the number of the fuzzy IF-THEN rules. The estimates for the un-measurable states and the adjustable parameters are obtained by using the weighted least squares estimator and its simplified one. It is proved that under some conditions the estimation errors will remain in the vicinity of zero as time increases, and the states are ultimately bounded subject to the proposed AFSMC. The effectiveness of the proposed method is indicated through the simulation experiment of a simple numerical system.     

基于Seam Carving的图像放大算法研究与改进

针对Seam Carving图像放大算法中,能量值前k小的像素带可能出现共用点的情况,提出了一种改进的图像放大自适应算法,通过判断图像中出现共用点的像素带的情况及计算其共用率的值,经过一定的策略确定待新增的像素带位置,该算法使得图像中视觉关注非重要信息区域像素带的增加能得到较均匀的分布;实现了人工干预选择视觉关注重要区域的方法,避免了自适应算法本身因误判视觉关注重要区域而带来关注对象发生变形扭曲的情况.实验证明,改进的算法能得到更好的放大效果.     

基于思维导图的团队协作系统

思维导图作为一种高效率的可视化工具,在线下团队协作中的运用也越来越多,而于在线协作工具的运用并不多. 该文针对思维导图在团队协作工具的应用来进行阐述,描述了一个采用B/S架构,以思维导图为重要功能模块的团队协作系统的实现. 阐述了其主要功能、系统结构、关键实现技术以及开发过程中相关的工程方法. 基于测试驱动的开发形式和IOC机制的运用,使其具备较好的扩展、维护和二次开发的能力. 而基于思维导图来展现团队协作中各类任务的相互关系,以图像化交互的形式来管理项目,与传统的协作系统的线性任务管理相比,具备更好的交互和较低的使用门槛.     

改进的协同过滤推荐算法

传统的选修课系统存在结构性的不足和缺憾,为了避免高校学生盲目的选择选修课程,本文利用改进的协同过滤算法对高校学生进行个性化的选课推荐.本文首先介绍了两种推荐算法,并着重介绍基于协同过滤的推荐算法,并分析了两种算法的优缺点,最后针对协同过滤算法的数据稀疏性问题,提出了一种改进的协同过滤算法,即在协同过滤中加入基于内容的因素来解决这个问题.这种改进的协同过滤算法避免了传统协同过滤算法中存在的数据稀疏问题,以学生为本推荐适合学生的课程,满足学生学习的个性化要求.     

A new robust double-talk detector based on the Stockwell transform for acoustic echo cancellation

Despite great developments in the field of acoustic echo cancellation (AEC), the presence of double-talk remains difficult problem. The main role of double-talk detection (DTD) is to control adaptation of the filter coefficients by halting their update in double-talk situations. In this paper, we propose a new method of DTD based on a time–frequency analysis that uses the Stockwell transform (ST).The ST is a time–frequency spectral localization method that combines the characteristics of the short-time Fourier transform and the wavelet transform. This method provides better time–frequency resolution, especially for non-stationary signals. In the experimental tests, the normalized least mean squares (NLMS) algorithm is used to update the filter coefficients along with speech signals taken from the TIMIT database. The obtained results show better performance compared to existing methods in terms of misalignment convergence and speech intelligibility enhancement.     

Similarity/dissimilarity calculation methods of DNA sequences: A survey

DNA sequence similarity/dissimilarity analysis is a fundamental task in computational biology, which is used to analyze the similarity of different DNA sequences for learning their evolutionary relationships. In past decades, a large number of similarity analysis methods for DNA sequence have been proposed due to the ever-growing demands. In order to learn the advances of DNA sequence similarity analysis, we make a survey and try to promote the development of this field. In this paper, we first introduce the related knowledge of DNA similarities analysis, including the data sets, similarities distance and output data. Then, we review recent algorithmic developments for DNA similarity analysis to represent a survey of the art in this field. At last, we summarize the corresponding tendencies and challenges in this research field. This survey concludes that although various DNA similarity analysis methods have been proposed, there still exist several further improvements or potential research directions in this field.