增量

Compared with average weighted and time-window algorithm, this paper introduce an attenuation factor to implement the user interest profile updating algorithm.

3对比平均加权算法、时间窗算法，引入遗忘因子，提出一种符合人的遗忘规律的增量式的用户兴趣模型更新算法。

Results The tinnitus frequency, severity of subjective tinnitus, audiogram types and some items of function test above hearing threshold were significantly different in the five groups (P.05). Specific features of pure-tone test were found in the five groups: tinnitus frequency was low, whole- frequency hearing threshold high, audiogram even, short increment sensitivity index at 1 kHz high and tone decline at 4 kHz low in the syndrome of invasion of wind-heat; tinnitus frequency was high, high-frequency hearing loss common, audiogram abruptly descending, SISI and TD at 4 kHz high in the syndrome of liver-fire up-stirring; subjective tinnitus was loud, audiogram abruptly descending or gradually descending, SISI and 7W at 4 kHz low in the syndrome of stagnation of phlegm-fire; tinnitus frequency was high, subjective tinnitus low, audiogram markedly descending or hill-like, SISI and TD at 4 kHz high in the syndrome of kidney-essence deficiency; average hearing loss was less, audiogram irregular but abruptly descending or notched in most cases, SISI at 4 kHz low in the syndrome of weakness of spleen-stomach.

结果］耳鸣频率、自觉耳鸣程度、听力图类型及部分阈上功能测试指标在各证型间具有显著性差异(P.05)，这些指标在不同的证型中表现出不同的特点：风热侵袭型的耳鸣频率较低，全频听阈升高，听力图多为平坦型，短增量敏感指数得分在1kHz较高，而4kHz音衰值较低；肝火上扰型的耳鸣频率较高，听力损失以高频为主，听力图多为陡降型，4kHz SISI得分及TD值均较高；痰火郁结型的自觉耳鸣响度较大，听力图以陡降型或缓降型为主，4kHz SISI得分及TD值均较低；肾精亏损型的耳鸣频率较高，而自觉耳鸣响度较低，听力图以显降型或山型为主，4kHz SISI得分及TD值均较高；脾胃虚弱型的平均听力损失较小，听力图无一定规律性，但以陡降型或切迹型为主，4kHz SISI得分较低。

Based on Hankel's matrix method and augmented least square recursive algorithm,the paper features the recursive least squares method of nonlinear parameter estimation with forgetting factor by introducing the constraint indicator of ...

依据Hankel矩阵法，在参数估计准则函数中加入待估参数的增量约束项和遗忘因子，并结合增广最小二乘递推算法，提出一种具有遗忘因子的非线性参数估计的递推最小二乘法。

The incremental equilibrium formulation and its equivalent iterative equation of the constant average acceleration Newmark method are deduced.

提出了增量形式Newmark法的等效迭代格式，该等效格式求解分缝拱坝非线性问题简便可行。

In the predicted process, the forward simplification is used to get the interface equation, meanwhile taking the voltage of the concentrated grid as control vector to solve the control step length; in the correction process, by means of forward simplification and back substitution the Newton linearization incremental equations of both concentrated computational grid and contracted nodes are solved alternately.

进行计算时，通过前向简化形成预估环节的接口方程，同时以浓缩网络的电压为控制电压向量求出控制步长；校正环节通过双向迭代方法交替实现对浓缩网络和收缩节点的牛顿法线性增量方程的求解。

The control volume integration was applied to deduce the discrete expressions of the convection-diffusion equations. The staggered grid and SIMPLE algorithm were introduced to deal with coupling between pressure and velocity, and then the numerical computation expressions of such variables as fluid flow velocity and pressure were deduced. Using the backward difference method and incremental theory to discretize the governing equations for fields of chemical reaction, material structure and chemorheology, the numerical computation expressions of variables such as the monomer conversion, average molecular weight and fluid viscosity were constructed.

应用控制容积积分法导出了对流—扩散方程的离散表达式，引入交错网格技术与SIMPLE算法，实现了耦合的压力场与速度场的分离式求解，推导得到了流体的流动速度、压力等物理量的数值计算式；采用向后差分方法和增量方法，实现了化学反应场、材料结构场、化学流变场控制方程的离散，获得了反应转化率、聚合物平均相对分子质量、流体黏度等物理量的数值计算式。

Baktun cycles having 400-years increment the secondary age category and synchronize the primary ages.

baktun周期有400年增量中学时代同步类初级佳话。

In addition,they should support incremental workow construction and must be able to run detached from the user console.

此外，它们必须能够支持增量式的工作流构建，而且必须能够以脱离用户控制的情况下运行。

The algorithm combines active learning, biased classification and incremental learning to model the small sample biased learning problem in relevance feedback process.

该算法将主动式学习、有偏分类和增量学习结合起来，对相关反馈过程中的小样本有偏学习问题进行建模。

This paper proposes an algorithm named Small Sample Incremental Biased Learning Algorithm to solve three difficulties of relevance feedback in sketch retrieval, including small sample issue, asymmetry of training data and real-time requirement.

为了解决草图检索相关反馈中小样本训练、数据不对称及实时性要求这3个难点问题，提出了一种小样本增量有偏学习算法。

Since historical times,England ,where the early inhabitants were Celts, has been conquered three times .

从有历史以来，英国，在此地早期居住的是凯尔特人，已经被征服了三次。

Bluetooth OBEX File Transfer Enables the sending and receiving of files on your phone via Bluetooth.

蓝牙OBEX文件移动允许经过蓝牙传送和接受文件。。。。