Knowledge-Driven Possibilistic Clustering with Automatic Cluster Elimination

Traditional Fuzzy C-Means(FCM)and Possibilistic C-Means(PCM)clustering algorithms are data-driven,and their objective function minimization process is based on the available numeric data.Recently,knowledge hints have been introduced to formknowledge-driven clustering algorithms,which reveal a data structure that considers not only the relationships between data but also the compatibility with knowledge hints.However,these algorithms cannot produce the optimal number of clusters by the clustering algorithm itself;they require the assistance of evaluation indices.Moreover,knowledge hints are usually used as part of the data structure(directly replacing some clustering centers),which severely limits the flexibility of the algorithm and can lead to knowledgemisguidance.To solve this problem,this study designs a newknowledge-driven clustering algorithmcalled the PCM clusteringwith High-density Points(HP-PCM),in which domain knowledge is represented in the form of so-called high-density points.First,a newdatadensitycalculation function is proposed.The Density Knowledge Points Extraction(DKPE)method is established to filter out high-density points from the dataset to form knowledge hints.Then,these hints are incorporated into the PCM objective function so that the clustering algorithm is guided by high-density points to discover the natural data structure.Finally,the initial number of clusters is set to be greater than the true one based on the number of knowledge hints.Then,the HP-PCM algorithm automatically determines the final number of clusters during the clustering process by considering the cluster elimination mechanism.Through experimental studies,including some comparative analyses,the results highlight the effectiveness of the proposed algorithm,such as the increased success rate in clustering,the ability to determine the optimal cluster number,and the faster convergence speed.

Metallic Softness Influence on Magic Numbers of Clusters

The metallic softness parameterαr 0 determines the structure of the cluster and governs the rule of magic numbers. Using molecular dynamic method, the stable structures and magic numbers are determined for the clusters consisting of 13 up to 147 atoms in medium range Morse potentials, which is suitable for most of metals. As the number of atoms constituting the cluster increases, the stable structures undergo transition from face-centered (FC) to edge-centered (EC) structures. The magic number take ones of FC series before transition and take ones of EC series after that. The transition point from FC to EC structures depends on the value of softness parameter.

Different Criteria for the Optimal Number of Clusters and Selection of Variables with R

One of the most important problems of clustering is to define the number of classes. In fact, it is not easy to find an appropriate method to measure whether the cluster configuration is acceptable or not. In this paper we propose a possible and non-automatic solution considering different criteria of clustering and comparing their results. In this way robust structures of an analyzed dataset can be often caught （or established） and an optimal cluster configuration, which presents a meaningful association, may be defined. In particular, we also focus on the variables which may be used in cluster analysis. In fact, variables which contain little clustering information can cause misleading and not-robustness results. Therefore, three algorithms are employed in this study： K-means partitioning methods, Partitioning Around Medoids （PAM） and the Heuristic Identification of Noisy Variables （HINoV）. The results are compared with robust methods ones.

基于非负矩阵分解的函数型聚类算法改进与比较

非负函数型数据可以不等间隔观测,在理论和实践中应用广泛,对其进行聚类可以更好地探索客观规律。文章利用位置积分变换将函数型数据转化为高维向量,再通过非负矩阵分解(NMF)将其转化为低维向量,以此构建函数型聚类算法。针对基于NMF的函数型谱聚类算法,给出了确定聚类个数K的两种方法:一种是根据Laplacian矩阵的特征值确定K;另一种是构建新评价指标,通过搜索确定K。数值实验结果显示:基于位置积分变换和NMF的函数型聚类算法有效,对函数结构要求宽松,但需限制函数取值为正;NMF的秩可通过cophenetic相关系数确定,建议取较小的值,以剔除类的冗余特征。在确定谱聚类的聚类个数K时,建议对降维后的数据进行标准化处理,以缩小样本间的距离变化范围;聚类个数变点图直观有效,再结合特征值差分法确定K很有参考价值,建议阈值取[0.05,0.08];根据吻合度与相似比确定K的方法有效且简单易懂。

Optimized air-ground data fusion method for mine slope modeling

Refined 3D modeling of mine slopes is pivotal for precise prediction of geological hazards.Aiming at the inadequacy of existing single modeling methods in comprehensively representing the overall and localized characteristics of mining slopes,this study introduces a new method that fuses model data from Unmanned aerial vehicles(UAV)tilt photogrammetry and 3D laser scanning through a data alignment algorithm based on control points.First,the mini batch K-Medoids algorithm is utilized to cluster the point cloud data from ground 3D laser scanning.Then,the elbow rule is applied to determine the optimal cluster number(K0),and the feature points are extracted.Next,the nearest neighbor point algorithm is employed to match the feature points obtained from UAV tilt photogrammetry,and the internal point coordinates are adjusted through the distanceweighted average to construct a 3D model.Finally,by integrating an engineering case study,the K0 value is determined to be 8,with a matching accuracy between the two model datasets ranging from 0.0669 to 1.0373 mm.Therefore,compared with the modeling method utilizing K-medoids clustering algorithm,the new modeling method significantly enhances the computational efficiency,the accuracy of selecting the optimal number of feature points in 3D laser scanning,and the precision of the 3D model derived from UAV tilt photogrammetry.This method provides a research foundation for constructing mine slope model.

一种基于属性相似性和分布结构连通性的聚类算法

聚类分析针对不同的数据特点采用不同的相似性度量,现实世界中数据分布复杂,存在分布无规律、密度不均匀等现象,单独考虑实例属性相似性或分布结构连通性会影响聚类效果。为此,提出了一种基于属性相似性和分布结构连通性的聚类算法(A Clustering Algorithm Based on Attribute Similarity and Distributed Structure Connectivity, ASDSC)。首先,利用待聚类数据集中的所有数据实例构建完全无向图,定义了一种兼顾属性相似和分布结构连通的新颖相似性度量方式,用于计算节点相似性,并构造邻接矩阵更新边的权重;其次,借助邻接矩阵执行递增步长的随机游走,依据顶点的连通中心性来识别簇中心并给定簇编号,同时获取其他顶点的连通性;然后,利用连通性计算顶点间的依赖关系,并据此进行簇编号的传播,直至完成聚类。最后,为了验证该方法的聚类性能,在16个合成数据集和10个真实数据集上与5种先进聚类算法进行了对比实验,ASDSC算法取得了优异性能。

改进的YoloV3轻量化多目标检测算法

为有效解决轻量化目标检测网络降低参数量导致的检测精度损失问题,设计一种轻量化多目标检测网络MSPF-YoloV3,可以通过低分组数、低内存占用的通道混洗结构,达到与高分组数同样的信息交流的效果;将位置信息丰富的浅层特征与深层特征进行融合,提高对不同大小目标的检测精度。经过实验验证了该方法的有效性与实时检测的能力。该网络在PASCAL VOC07+12数据集上的mAP达到86.31%;在COCO2014数据集中提取出与PASCAL VOC07+12数据集相同的20个类别,mAP达到67.71%;权重文件大小为46.8 MB,相较于YoloV3降低了198.2 MB;检测速度达到44 FPS。

基于HS-Clustering的风电场机组分组功率预测

为了寻求风电场功率预测精度和计算效率二者的平衡,提出了一种基于霍普金斯统计量与聚类算法(HSClustering)的风电场机组分组功率预测方法,该方法将霍普金斯统计量与聚类算法的优势有效结合,采用霍普金斯统计量确定场内机组分组个数,通过聚类算法识别不同机组的相似性将风电场分成不同的机组群,然后对每组机群分别建立功率预测模型,从而叠加得到整场输出功率;另外以实测风速、实测功率及二者组合作为机组分组模型输入,分析其对预测精度的影响程度。实例分析表明基于HSClustering的分组预测方法可以显著提高预测精度,同时保证较高的计算效率;风速是影响分组效果的主要因素,对于某些分组模型,功率又可以作为风速的重要补充。

基于改进K-means算法的大跨屋盖结构表面风荷载分区研究

针对K-means算法进行大跨屋盖结构表面风荷载分区中存在的分类数k值需凭经验事先给定以及所有初始聚类中心均需随机选取带来的分类情况数过多、从中寻找最优分类结果工作量大且效率低的问题,提出基于改进K-means算法的大跨屋盖结构表面风荷载分区方法。首先,建立分类数k与其相应测点风荷载的误差平方和(Sum of the Squared Errors:SSE)关系曲线,引入手肘法基本思想,实现最优分类数kst值的精准识别;其次,在首个初始聚类中心随机选取基础上,引入轮盘法基本思想,完成对剩余初始聚类中心的高效选取;然后,根据类内紧凑、类间分散的原则,通过类内紧凑性判定指标S(k)和类间分散性判定指标D(k),构造并借助SD(k)值有效性检验,得到最优的风荷载分区结果;最后,以北京奥林匹克网球中心大跨悬挑屋盖结构为例,针对风洞试验所得风荷载测试结果,采用所提方法对其表面最不利风压系数进行分区计算,并与传统K-means算法进行对比,结果表明,所提方法能够高效实现大跨屋盖结构表面风压分区计算,具有较好的工程应用价值。

自适应邻域密度聚类及事故黑点识别应用

聚类作为识别交通事故黑点的主要方法之一,其主要问题是交通事故多发区事先无法确定,即无法提前知道聚类簇数。利用样本点之间的连接概率定义了数据点的局部密度,根据局部密度大小来确定聚类中心和簇数,再对数据点进行聚类。结果表明:一是算法对参数不敏感,具有较好的通用性;二是算法能自动确定聚类簇数;三是算法聚类过程只依赖局部密度与邻接点,能够识别噪声点,提升结果的准确性。运用算法在一些真实数据集上进行试验,将聚类结果与其他算法结果利用评价指标ARI(Adjusted Rand Index)和NMI(Normalized Mutual Information)进行比较。最后利用算法对美国6个州的交通事故进行聚类,结果表明算法对交通事故有较好的适应性,能将城市及周边道路上事故密集区域准确识别出来。

基于CiteSpace软件的全球土壤改良剂研究进展

土壤改良剂不给作物提供养分,但可以改善土壤性质,促进作物吸收养分。收集2000-2023年发表在Webof Science数据库核心合集的9360篇文献,采用CiteSpace软件作为分析工具,简述全球土壤改良剂发文情况,开展关键词聚类分析,展望全球重点研究方向。结果表明,我国在土壤改良剂方面有着强大的科研能力和影响力,发文量远超其他国家。全球土壤改良剂研究的发文量增长迅速,研究范畴越来越广泛。未来,要充分利用工农业废弃物,研发新型土壤改良剂,混合施用多种改良剂,从而有效改善土壤性质,促进农业生产。

A Study on Numerical Calculation Method of Small Cluster Density in Percolation Model

Percolation theory deals with the numbers and properties of the clusters formed in the different occupation probability. In this Paper, we study the calculation method of small clusters. We calcu-lated the small cluster density of 1, 2 and 3 in the percolation model with the exact method and the numerical method. The results of the two methods are very close, which can be verified by each other. We find that the cluster density of all three kinds of small clusters reaches the highest value when the occupation probability is between 0.1 and 0.2. It is very difficult to get the analytical formula for the exact method when the cluster area is relatively large (such as the area is more than 50), so we can get the density value of the cluster by numerical method. We find that the time required calculating the cluster density is proportional to the percolation area, which is indepen-dent of the cluster size and the occupation probability.

页岩气井出砂主控因素分析及防治对策

随着地层压力逐渐降低,川渝地区页岩气田加密调整时,压裂返排出现支撑剂回流、返吐现象,一方面降低支撑裂缝的导流能力,另一方面支撑剂进入地面给设备带来冲刷损坏。为从源头上控制地层出砂,开展主控因素分析,为页岩气井立体防砂措施提供参考依据。通过改善支撑剂的铺置剖面,增加压裂液粘度、优化单段射孔簇数、加入纤维等优化措施,有效降低支撑剂回流风险。并对放喷测试制度进行优化调整,采用由小到大的油嘴顺序进行放喷。并针对不同程度出砂气井,形成“一井一策”差异化治理措施,提出旋流过滤式除砂器工艺,除砂器纵向上整体采用“内窄外宽”设计,有效避免堵塞缝腔,对粒径≥0.6 mm粉砂除砂效率达到98%。

2016—2021年5—10月辽宁省降水量时空分布精细化特征分析

选用2016—2021年5—10月辽宁省小时加密观测降水资料,分析辽宁省降水过程、短时强降水过程、极端短时强降水过程降水量时空分布特征,应用K均值聚类法对汛期短时强降水过程降水累积小时数的空间分布进行分类。结果表明:辽宁省短时强降水过程(小时降水量大于或等于20 mm)和极端短时强降水过程(小时降水量大于或等于该站点历史降水量升序后的第99.9百分位)受地形影响较大。短时强降水高发区主要分布在辽宁东南部的沿海地区、辽宁西部的沿海平原和地势较高地区及辽宁东南部的迎风坡地区。辽宁省汛期降水受季风环流影响,有明显的季节变化,平均降水量和降水频次最大值发生在7月和8月,降水时段主要集中在午后至夜间。辽宁省短时强降水过程降水频次的区域间差异较大,极端短时强降水过程区域间差异较小。

无线传感网LEACH协议的改进方法

为解决传统LEACH(Low Energy Adaptive Clustering Hierarchy)协议网络节点能量消耗高、存活数量少和生存寿命短等问题,提出了一种LEACH-AD改进方案。该算法引入最优簇头比率P值、加入距离因子、剩余能量因子和密度因子等因素更新的阈值公式进行分簇以及簇间的传输。实验结果表明,改进后的LEACH-AD协议在首个死亡节点、10%死亡节点以及全部死亡节点分别比原LEACH协议延长138轮、195轮、628轮。在能量消耗方面比原LEACH协议多持续了631轮,改进后的路由协议减少了网络节点的能量消耗量,从而有效延长了无线网络与传感节点的工作时间,这对无线监测系统的研究与开发意义重大。

The Effective Clustering Partition Algorithm Based on the Genetic Evolution

To the problem that it is hard to determine the clustering number and the abnormal points by using the clustering validity function, an effective clustering partition model based on the genetic algorithm is built in this paper. The solution to the problem is formed by the combination of the clustering partition and the encoding samples, and the fitness function is defined by the distances among and within clusters. The clustering number and the samples in each cluster are determined and the abnormal points are distinguished by implementing the triple random crossover operator and the mutation. Based on the known sample data, the results of the novel method and the clustering validity function are compared. Numerical experiments are given and the results show that the novel method is more effective.

Analysis of the Employment Situation of Non Private Enterprises in Various Regions of China

In the past 30 years, Chinese enterprises have been a hot topic of discussion and concern among the general public in terms of economic and social status, ownership structure, business mechanism, and management level. Solving the problem of employment for the people is an important prerequisite for their peaceful living and work, as well as a prerequisite and foundation for building a harmonious society. The employment situation of private enterprises has always been of great concern to the outside world, and these two major jobs have always occupied an important position in the employment field of China that cannot be ignored. With the establishment of the market economy system, individual and private enterprises have become important components of the socialist economy, making significant contributions to economic development and social progress. The rapid development of China’s economy, on the one hand, is the embodiment of the superiority of China’s socialist market economic system, and on the other hand, it is the role of the tertiary industry and private enterprises in promoting the national economy. Since the 1990s, China’s private enterprises have become a new economic growth point for local and even national countries, and are one of the important ways to arrange employment and achieve social stability. This paper studies the employment of private enterprises and individuals from the perspective of statistics, extracts relevant data from China statistical Yearbook, uses the relevant knowledge of statistics to process the data, obtains the conclusion and puts forward relevant constructive suggestions.

A numerical model for cloud cavitation based on bubble cluster

The cavitation cloud of different internal structures results in different collapse pressures owing to the interaction among bubbles. The internal structure of cloud cavitation is required to accurately predict collapse pressure. A cavitation model was developed through dimensional analysis and direct numerical simulation of collapse of bubble cluster. Bubble number density was included in proposed model to characterize the internal structure of bubble cloud. Implemented on flows over a projectile, the proposed model predicts a higher collapse pressure compared with Singhal model. Results indicate that the collapse pressure of detached cavitation cloud is affected by bubble number density.

基于CiteSpace的国内患者安全教育研究现状的文献计量学分析

目的:采用CiteSpace软件对我国患者安全教育相关文献进行文献计量学分析,总结该领域研究热点,探索未来发展趋势。方法:检索中国知网建库至2023年7月25日发表的相关文献,使用文献计量学分析方法及CiteSpace软件进行统计分析。结果:共检索出163篇文献,发文量呈现逐年上升的趋势。发文量居前3位的机构分别为重庆医科大学附属第一医院、天津医科大学和华中科技大学同济医学院附属协和医院。研究热点主要包括安全教育、调查、新护士、安全管理、安全文化及医学生6个聚类。影响因素、护士和新护士成为近3年的研究热词,并呈现持续上升趋势。结论:国内患者安全教育的发文量呈现增长趋势,发文机构以高等医学院校及附属医院为主,作者间的合作较弱,鉴于患者安全教育对临床医疗质量与安全的重要性,未来研究者应给予重视,加强机构间的合作与交流,就患者安全教育方案进行深入探讨。

无人机集群编队自主协同控制方法综述

自主协同控制是实现无人机集群高效协同编队的一种主流控制方法,属于当前无人机控制领域的研究热点和难点。论文首先从两类无人机集群编队自主协同控制主流类型切入。然后,重点对集中式控制类型中的领航者-跟随者法、虚拟结构法,和分布式控制类型中的人工势场法、行为法、一致性理论法、深度强化学习法等典型方法的基本思想、特点和最新研究进展进行了对比分析。最后,对无人机集群编队自主协同控制方法的发展趋势进行展望,以期为后续研究提供参考。