基于蚁群优化算法和k阶近邻法的暂态稳定评估特征选择

提出了基于蚁群优化算法和k阶近邻法相结合的嵌入式特征选择算法.选择稳态潮流量构成电力系统暂态稳定评估的输入特征集,针对输入特征集包含的大量冗余信息,特征选择结果中可能包含一定冗余特征的缺陷,先用聚类的方法裁剪冗余性特征,然后用所提算法选择和稳定状况强相关的关键特征,提高了特征选择的 效率.通过对3机9节点和10机...     

A Dynamic k-Nearest Neighbor Method for WLAN-Based Positioning Systems

The static k-Nearest Neighbor (k-NN) method for localization has limitations in accuracy due to the fixed k value in the algorithm. To address this problem, and achieve better accuracy, we propose a new dynamic k-Nearest Neighbor (Dk-NN) method in which the optimal k value changes based on the topologies and distances of its nearest neighbors. The proposed method has been validated using the WLAN-fingerprint data sets collected at COEX, one of the largest convention centers in Seoul, Korea. The proposed method significantly reduced both the mean error distances and the standard deviations of location estimations, leading to a significant improvement in accuracy by ~ 23% compared to the cluster filtered k-NN (CFK) method, and ~ 17% compared to the k-NN (k = 1) method.     

一种k-NN分类器k值自动选取方法

k-NN 分类算法已广泛应用于文本挖掘和模式识别等领域, 其近邻数k直接影响着分类精度, k 值过小
时k-NN 会受到噪声的影响, k值过大时同样会降低分类精度, 为此提出一种快速选取k值的方法. 首先给出k 值的
候选集, 然后在候选集上快速地选取k 值. 在100 个公开数据集上的实验结果表明, 所提出的算法能够选取一个有效
的近邻数k 是一种效果好、有潜力的方法.      

基于最短路径的道路网络k近邻查询处理

针对基于空间道路网络的k近部查询处理,提出了分布式移动对象更新策略以有效减少服务器计算代价,利用基于内存的空间道路网络部接矩阵、最短路径矩阵结构和移动对象哈希表索引分别对道路网络无向图与移动对象进行存储管理。提出了基于最短路径度量的网络扩展搜索(SPNE)算法,以通过裁剪网络搜索空间来减少k近部查询搜索代价。实验表明,SPNE算法的性能优于传统的NE和MKNN等k近邻查询处理算法。     

Particle swarm optimization for ensembling generation for evidential k-nearest-neighbour classifier

The problem addressed in this paper concerns the ensembling generation for evidential k-nearest-neighbour classifier. An efficient method based on particle swarm optimization (PSO) is here proposed. We improve the performance of the evidential k-nearest-neighbour (EkNN) classifier using a random subspace based ensembling method. Given a set of random subspace EkNN classifier, a PSO is used for obtaining the best parameters of the set of evidential k-nearest-neighbour classifiers, finally these classifiers are combined by the “vote rule”. The performance improvement with respect to the state-of-the-art approaches is validated through experiments with several benchmark datasets.

基于TBM的自适应模糊k-NN分类器

针对训练模式类标签不精确的识别问题,提出基于可传递信度模型的自适应模糊k-NN（k-Nearest Neighbor）分类器。利用可传递信度模型结合模糊集理论和可能性理论并运用pignistic变换,对待识别模式真正所属的类做出决策。采用梯度下降最小化误差函数,以实现参数的自适应学习。实验结果表明,该分类器误分类率低、鲁棒性强。     

Nearest neighbor imputation of species-level, plot-scale forest structure attributes from LiDAR data

Meaningful relationships between forest structure attributes measured in representative field plots on the ground and remotely sensed data measured comprehensively across the same forested landscape facilitate the production of maps of forest attributes such as basal area (BA) and tree density (TD). Because imputation methods can efficiently predict multiple response variables simultaneously, they may be usefully applied to map several structural attributes at the species-level. We compared several approaches for imputing the response variables BA and TD, aggregated at the plot-scale and species-level, from topographic and canopy structure predictor variables derived from discrete-return airborne LiDAR data. The predictor and response variables were associated using imputation techniques based on normalized and unnormalized Euclidean distance, Mahalanobis distance, Independent Component Analysis (ICA), Canonical Correlation Analysis (aka Most Similar Neighbor, or MSN), Canonical Correspondence Analysis (aka Gradient Nearest Neighbor, or GNN), and Random Forest (RF). To compare and evaluate these approaches, we computed a scaled Root Mean Square Distance (RMSD) between observed and imputed plot-level BA and TD for 11 conifer species sampled in north-central Idaho. We found that RF produced the best results overall, especially after reducing the number of response variables to the most important species in each plot with regard to BA and TD. We concluded that RF was the most robust and flexible among the imputation methods we tested. We also concluded that canopy structure and topographic metrics derived from LiDAR surveys can be very useful for species-level imputation.     

Investigation of wave atom transform by using the classification of mammograms

This paper presents an approach for breast cancer diagnosis in digital mammograms using wave atom transform. Wave atom is a recent member of the multi-resolution representation methods. Primarily, the mammogram images are decomposed on the basis of wave atoms, and then a special set of the biggest coefficients from wave atom transform is used as a feature vector. Two different classifiers, support vector machine and k-nearest neighbors, are employed to classify mammograms. The method is tested using two different sets of images provided by MIAS and DDSM database.     

CBR灾害救助系统中一种权重调整算法

针对灾害救助系统中的案例检索问题,结合经典案例相似度计算方法,对目前应用广泛的k-NN算法进行改进。在假设时间因素对历史案例可采纳程度有显著影响的基础上,提出了一种基于时序的案例属性权重调整算法,并在实验中证明了该改进算法的有效性。