This thesis discusses how to acquire knowledge in incomplete information system by bayes theory , d - s evidence theory , analyse the relation and difference of them , and illuminate it by example 本文探讨了证据理论、贝叶斯理论在不完备信息系统中如何进行规则提取，分析了它们两者之间的联系与区别，并通过实例加以说明。
Evidence theory is . a kind of important reasoning method , dempster first puts it forward in 1960 ' s , it wants to use probability upper and lower limit to express the uncertainty in practical problem 证据理论是dempster于20世纪60年代首次提出，试图用概率上下限来表示实际问题中的不确定性的一种重要的不确定性推理方法。
Based on former information filtering approaches and the i ) - s theory , this thesis gains cop " s profile by model ing the interest of cop , studies and implements the cop orientated information filtering technique 基于已有的信息过滤研究与证据理论，本文对cop的兴趣进行建模，得到cop的兴趣特征，并以此为基础研究和实现了面向cop的信息过滤技术。
At last , through comparing each fusion method , this paper presents a kind mixed configuration fusion algorithm based on rough sets theory 、 neural network and d - s evidence theory , and uses this method in target recognition 最后在比较各种融合方法的基础上，提出了采用基于粗集理论、神经网络和证据理论的混合结构的融合算法，并将其应用在目标识别中。
According to the target recognition results in remote sensing images and the target states estimation results predicted by the kalman filter , a new robust target association algorithm based on d - s evidence theory was proposed 摘要以遥感图像目标识别结果为主，结合卡尔曼滤波器获得的目标运动状态估计结果，提出一种基于d - s证据理论的目标数据联合关联算法。
Based on the classification results , to improve the classification veracity of multi - band sar system , this paper proposes a decision fusion method based on dempster - shafer evidence theory to fuse the classification results 在图像分类的基础上,为了提高多波段sar图像分类准确性,本文提出了基于dempster - shafer证据理论的多波段sar图像分类结果的决策级融合。
When the discount coefficient is 1 and all weights of the nearest neighbor sample points are the same , the k - nn classification method based on evidence reasoning model will become the k - nn classification method based on evidence theory 并且当折扣系数为1 ，且给定所有最近邻样本点权重相等时，基于证据推理模型的k - nn分类方法就成为基于证据理论的k - nn分类方法。
In evidence theory , it is an important that how to obtain basic probability assignment ( bpa ) function . according to the classification theory of remote sensing image , bpa function and the rule of fusion classification are determined respectively 根据遥感图像分类原理以及证据理论的特点，分别确定了bdset方法和fdset方法的基本概率分配函数及融合分类规则。
Back - propagation ( bp ) neural networks are used as classification models in various feature domains , and then outputs of these networks are combined by the d - s evidence theory to obtain a consensus classification re suit 待辨识的6类手部动作肌电信号经各特徵域变换，提取特徵矢量后由bp神经网络分类，根据d - s证据理论对各分类器分类结果进行证据累积，并得到最终分类结果。
From the viewpoint of discrimination and using d - s evidence theory , this method realized the management of sensor resource under the optimization rules to maximize the forecasted recognition discrimination under the restrictions 该方法从分辨力的角度出发，利用d - s证据理论进行推理，在满足约束的条件下，依据使系统预测识别分辨力最大的原则进行优化，实现传感器资源的管理。
