New Algorithms for Data Discretization Based on Rough Set Theory

The discretization of real values is always one of the key problems to be solved in the domain of machine learning for its great contribution to speeding up the followed learning algorithms, cutting down the real demand of algorithms on running space and time, and improving the clustering capability of the ultimate learning results. The basic characteristics and framework of discretization approaches based on rough set model are analyzed at first, then the different measurements of the importance of candidate cuts are discussed and researched. Two new heuristic algorithms are put forward to finally select the useful cuts from a candidate set. The selected cuts of the two algorithms will adequately maintain the discernible relation of information systems for their full considering the specialty of rough set, which perfectly embodies the advantages of this theory. Moreover, excellent discretization results may be expected through these heuristic algorithms.     

On System Stability under Random Perturbations

In this paper we concider the problem of stability of stochastic evolution systemsin Hilbert space drived by a cylindrical Brownian mothin. We regard the stochestic evolutionequation dXt = AXtdt +G(Xt)dBt as a deterministic system of the form dXt=AXtdt under randomperturbation,and obtain stability of its solution. It is shown that under certain assumptions,itsevolution solution and L2-contimuous evulution solution are exponertially stable.     

Asymptotic Attractors of the Nonlinear Evolution Equation in Bounded Domain

The basic principle of infinite-dimensional dynamic system is to try to reduce the original infinite-dimensional system to an infinite-dimensional system. However,due to the unknown structure of the reduced system, it is difficult to describe its dynamical behaviour. To overcome this difficulty, the idea of approximate inertial manifolds is introduced, for NSE, the existence of AIM was studied, it is shown that the global attractor lies within a neighborhood of the graph of an Lipschitz function by the squeezing property. In this paper, by constructing a finite dimensional solution sequence, we will prove that it tends to the global attractor, theoretically, this provides a metod of constructing the asymptotic attractors, theoretically, this provides a method of constructing the asymptotic attractors for the evolution equations.     

THE INDUSTRIAL ORGANIZATIONS AND THE L-S-BUSINESS STRATEGIES

This papar presents an evolution model in business behaviors-the L-S Business Strategies,gives some examples and predictions which support these strategies,and analyzes the relationship between the industrial organizations in China and the L-S Business strategies.     

面粉品质性状与速冻水饺品质关系的研究

对7种面粉的化学成分、面团流变学特性等指标进行测定,详细分析了各项指标与速冻水饺品质的相关关系,得出适合生产速冻水饺的面粉的最佳特性指标。并通过比较分析对黑龙江省地产小麦粉生产速冻水饺的适用性进行评价。     

如何建立成本分析模板服务一拖公司的产品结构调整

介绍一拖公司财务部在传统的成本分析理论基础上,建立了成本因素分析方法,为各单位产品成本管理和决策工作提供及时、快速的分析工具,并协助各单位找出影响利润的关键点,以改善盈利能力,提升产品竞争力;同时,建立完善了产品谱系盈利分析模板,引导一拖公司通过产品结构调整,提升集团公司的创利能力。     

基于分解机制的多目标蝙蝠算法

在分析蝙蝠算法性能基础上,将蝙蝠算法融入分解机制,提出了一种基于分解机制的多目标蝙蝠算法。为了进一步提高算法的多样性,将差分进化策略引入算法中。对14个具有复杂Pareto前沿的多目标优化问题(LZ-09系列和ZDT系列)测试不同邻域规模对算法性能的影响,结果表明新算法的邻域规模为20时性能最优;将其与MOEA/D-DE和NSGA-II算法进行对比分析,结果显示该算法的分布性、收敛性和多样性均优于另外两种算法。为了验证其求解含有约束问题的性能,将其应用于滑动轴承多目标优化设计问题中,获得的Pareto前沿分布均匀,表明算法具有工程实用性,是求解复杂高维多目标问题的有效方法。