模型理论

Among used machine learning methods, the gradient descent method is widely used to train various classifiers, such as Back-propagation neural network and linear text classifier. However, the gradient descent method is easily trapped into a local minimum and slowly converges. Thus, this study presents a gradient forecasting search method based on prediction methods to enhance the performance of the gradient descent method in order to develop a more efficient and precise machine learning method for Web mining.However, a prediction method with few sample data items and precise forecasting ability is a key issue to the gradient forecasting search method. Applying statistic-based prediction methods to implement GFSM is unsuitable because they require a large number of data items to model a prediction model. In the contrast with statistic-based prediction methods, GM(1,1) grey prediction model does not need a large number of data items to build a prediction model, and it has low computational load. However, the original GM(1,1) grey prediction model uses a mathematical hypothesis and approximation to transform a continuous differential equation into a discrete difference equation in order to model a forecasting model.

其中梯度法是一个最常被使用来实现机器学习的方法之一，然而梯度法具有学习速度慢以及容易陷入局部最佳解的缺点，因此，本研究提出一个梯度预测搜寻法则(gradient forecasting search method， GFSM)来改善传统梯度法的缺点，用来提升一些以梯度学习法则为基础的分类器在资讯探勘上的效率与正确性；而一个所需资料量少、计算复杂度低且精确的预测模型是梯度预测搜寻法能否有效进行最佳解搜寻之关键因素，传统统计为基础之预测方法的缺点是需要较大量的数据进行预测，因此计算复杂度高，灰色预测模型具有建模资料少且计算复杂度低等优点，然而灰色预测理论以连续之微分方程式为基础，并且透过一些数学上的假设与近似，将连续之微分方程式转换成离散之差分方程式来对离散型资料进行建模及预测，这样的作法不尽合理，且缺乏数学理论上的完备性，因为在转换过程中已经造成建模上的误差，且建模过程仅考虑相邻的两个资料点关系，无法正确反应数列未来的变化趋势。

Based on the research work of basic theories including phase space reconstruction,embedding theorem,correlation dimension,local dynamics,Lyapunov exponents,surrogate data etc,based on the research work of general methods such as principal component analysis,correlation dimension GP algorithm,false neighbors method,nonlinear time series prediction,local prediction, adaptive prediction,neural network model,support vector machines regression model, prediction power,nonlinear detection,coarse-graining methodology,conditional entropy and so on,the framework of nonlinear time series analysis are constructed.

在对包括相空间重构、嵌入定理、关联维数、局部动力学、Lyapunov指数、替代数据(来源：A97BC论文网www.abclunwen.com)、等基本理论与其物理意义的研究和讨论基础上；在对包括主分量分析、关联维数GP算法、伪邻近点法、非线性时间序列预测、局域预测、自适应预测、神经网络模型、支持向量回归模型、预测效果、非线性检测、粗粒化方法、条件熵等非线性时间序列一般分析方法的原理和算法研究基础上；构建了新的非线性时间序列分析的理论体系，归纳总结了非线性时间序列分析的基本问题和主要研究方面。2。

The reliability theory is combined with the fuzzy optimization technique, the theories and methods on the symmetrical fuzzy reliability optimization design, the asymmetric fuzzy reliability optimization design and the multi-objective fuzzy reliability optimization design are presented respectively, and then the correspondence direct model solutions, the iterative solutions, the corrected algorithm for iteration method, the level cuts solutions and the searching solutions with limited boundary for the asymmetric fuzzy reliability design are presented respectively.

将可靠性理论与模糊优化技术有机地结合起来，提出了结构对称模糊可靠性优化设计、非对称模糊可靠性优化设计和多目标模糊可靠性优化设计理论与方法，并分别给出了模型的直接解法、迭代解法、迭代求解的改进算法和非对称模糊可靠性优化设计模型的水平截集法和限界搜索法。

Based on a nonclassical theory of plasticity and the continuum damage mechanics,a damage constitutive equation is obtained for the ferrite phase with the damage evolution law for mixed hardening materials from an ellipsoidal void model under particular coordinates,where the void shape variation is considered.

基于非经典塑性理论和连续介质损伤力学，利用在一个特殊坐标系下基于椭球形孔洞模型得到的可考虑孔洞形状变化混合强化材料的损伤演化率得到了铁素体相的损伤本构方程，通过混合物理论利用铁素体和渗碳体相各自本构关系并考虑其几何特征得到了珠光体团的损伤本构模型。

The model in this paper takes the form as belowThe main contents and results in this paper are as follows:In section 1, we state our main analytical results on equilibria of the above system. First, several sufficient conditions for coexistence of the steady-state are given by the standard fixed-point index theory in cone. Second, the global structure of the coexistence solutions and their local stability are established by using bifurcation theory. Third, the multiplicity, uniqueness and stability of positive steady-state solutions to this system are derived by means of perturbation theory of eigenvalues, standard regularity theory, Sobolev embedding theorem and fixed-point index theory.

本文主要有两章内容：第一章研究了该模型正平衡解的性质，可分为三部分：第一部分运用极值原理、上下解方法和锥映射不动点指标理论得到正平衡解存在的充分条件；第二部分利用分歧理论给出了平衡态系统正分歧解的结构，并讨论了局部分歧解的稳定性；第三部分利用特征值扰动理论、标准椭圆正则化理论、Sobolev嵌入定理及指标理论讨论了正平衡解的多重性、唯一性及稳定性，并讨论了参数对解的影响。

From this perspective, the dissertation propounds theoretic framework of Hainan CGE model, prepares Hainan Social Accounting Matrix integrating Hainan's statistic data, forms an equilibrious and uniform data set, and on this basis constructs a CGE model referring to Hainan's six departments to next study impact of policy changes about exemption of agricultural tax for specific produces and further reduction of agricultural tax on Hainan's economic structure, producer welfare, consumer welfare, fiscal revenue and so forth, and provide certain theory and means for enaction, analysis and evaluation of provincial policy.

本文从这个角度出发，提出了海南省CGE模型的理论框架，结合海南省统计数据的实际情况，编制出海南省社会核算矩阵，形成一个均衡一致的数据集，并在此基础上构建出一个海南省六部门的CGE模型，用于下一步研究取消农业特产税和进一步降低农业税等政策变化对海南经济结构、生产者福利、消费者福利、财政收入等方面的影响，为省级政策的制定、分析和评价提供了一定的理论和方法支撑。

The first stage is an endothermic reaction, and the second one is an exothermic reaction.

采用双组分分阶段一级反应模型求出了四个污泥样品的表观活化能和指前因子，给出了样品热解的动力学方程，通过理论计算，双组分分阶段一级反应模型的理论数据与实验数据吻合良好。

In this dissertation, the author attempts to apply the recent researches on metaphor -- the "two-domain" model from the experientialist theory and the "many-space" model from the conceptual integration theory to the translation of metaphor in Chinese classical poetry.

在本文中，作者尝试应用近年来对隐喻的研究成果－包括体验主义理论的"双域"模型和概念融合理论的"多空间"模型―来探索理解中国古典诗歌中的隐喻，并付诸翻译过程中。

In this dissertation, the author attempts to apply the recent researches on metaphor-the "two-domain" model from the experientialist theory and the "many-space" model from the conceptual integration theory to the translation of metaphor in Chinese classical poetry.This paper classifies metaphor into four types with LA. Richards terms.

在本文中，作者尝试应用近年来对隐喻的研究成果—包括体验主义理论的"双域"模型和概念融合理论的"多空间"模型—来探索理解中国古典诗歌中的隐喻，并付诸翻译过程中。

Through detailed analyzing some influencing factiors of sand liquefaction, seven parameters were selected as assessment indexes. They are earthquake magnitude, peak ground surface acceleration, standard penetration value, specific penetration resistance, relative compaction, average particle diameter, and water table. The extension assessment model has been established that was applied to assessment of sand liquefaction potential. The model had been trained by the historical data and has been assessed in virtue of sand liquefaction from the historical data by simulating, and got a excellent result. After introduction of the support vector machine theory, this paper established the assessment model of sand liquefaction based on the support vector machine, and assessed in sand liquefaction from virtue data of some fields in home and overseas with the tool Libsvm, which was programmed based on the support vector machine theory. And got a good coherency between the assessement results and the actual liquefaction.

在详细分析影响砂土液化因素的基础上，选取了震级M、地面最大加速度g_、标准贯入击数N_(63.5)、比贯入阻力P_s、相对密实度D_r、平均粒径D_(50)、地下水位d_w等7个参数作为评价指标，建立了砂土液化势的可拓评价模型，并对历史数据进行仿真评价，取得了较好的应用效果；在对支持向量机的理论进行介绍后，本文建立了基于支持向量机的砂土液化势评价模型，并利用在基于支持向量机理论的Libsvm软件，对若干国内外场地的现场实测数据进行了评价，评价结果与现场表现基本上一致。

I didn't watch TV last night, because it .

昨晚我没有看电视，因为电视机坏了。

Since this year, in a lot of villages of Beijing, TV of elevator liquid crystal was removed.

今年以来，在北京的很多小区里，电梯液晶电视被撤了下来。