Changepoint Detection with Outliers Based on RWPCA

Changepoint detection faces challenges when outlier data are present. This paper proposes a multivariate changepoint detection method which is based on the robust WPCA projection direction and the robust RFPOP method, RWPCA-RFPOP method. Our method is double robust which is suitable for detecting mean changepoints in multivariate normal data with high correlations between variables that include outliers. Simulation results demonstrate that our method provides strong guarantees on both the number and location of changepoints in the presence of outliers. Finally, our method is well applied in an ACGH dataset.

Industrial Spatial Agglomeration Using Distance-based Approach in Beijing, China

To study the difference of industrial location among different industries, this article is to test the spatial agglomeration across industries and firm sizes at the city level. Our research bases on a unique plant-level data set of Beijing and employs a distance-based approach, which considers space as continuous. Unlike previous studies, we set two sets of references for service and manufacturing industries respectively to adapt to the investigation in the intra-urban area. Comparing among eight types of industries and different firm sizes, we find that： 1） producer service, high-tech industries and labor-intensive manufacturing industries are more likely to cluster, whereas personal service and capital-intensive industries tend to be randomly dispersed in Beijing; 2） the spillover of the co-location of finns is more important to knowledge-intensive industries and has more significant impact on their allocation than business-oriented services in the intra-urban area; 3） the spatial agglomeration of service industries are driven by larger establishments, whereas manufac- turing industries are mixed.

Comparison of Analyses of Genetic Structure among Chinese Indigenous Chicken Breeds using Distance-based and Model-based Methods

The Nei's improved genetic distance(DA)and gene flow(Nm)were measured using sixteen microsatellite markers.Dendograms based on DA genetic distance using the neighbor-joining(NJ)method and STRUCTURE program were constructed to analyze the genetic structure and relationship among 10 Chinese indigenous chicken breeds.The results showed that dendograms of DA genetic distance using the NJ method divided the 10 chicken breeds into two main clusters;one consisted of breeds of low weight body(CHA,TTB,XIA,GUS and BAI),the other contained heavier breeds(LAN,DAG,YOU,XIS and LUY).In the lighter breeds,TIB and CHA clustered together,as did XIA and GUS.In the heavier breeds,XIS and LUY was clustered together in one branch,but LAN,DAG and YOU clustered in independent branches.The results were consistent with Nm estimates among the 10 indigenous chicken breeds.The STRUCTURE program properly inferred the presence of genetic structure despite not pre-defining the origin of individuals.The genetic cluster inferred by STRUCTURE was basically the same as that from the DA distance clustering method.An advantage of the STRUCTURE program was its ability to identify the migrants and admixed individuals in the 10 chicken populations;this could not be achieved by use of the DA distance clustering method.

A Novel Method for Prediction of Protein Domain Using Distance-Based Maximal Entropy

Detecting the boundaries of protein domains is an important and challenging task in both experimental and computational structural biology. In this paper, a promising method for detecting the domain structure of a protein from sequence information alone is presented. The method is based on analyzing multiple sequence alignments derived from a database search. Multiple measures are defined to quantify the domain information content of each position along the sequence. Then they are combined into a single predictor using support vector machine. What is more important, the domain detection is first taken as an imbal- anced data learning problem. A novel undersampling method is proposed on distance-based maximal entropy in the feature space of Support Vector Machine （SVM）. The overall precision is about 80%. Simulation results demonstrate that the method can help not only in predicting the complete 3D structure of a protein but also in the machine learning system on general im- balanced datasets.

Bi-Level Programming for the Optimal Nonlinear Distance-Based Transit Fare Structure Incorporating Principal-Agent Game

The urban transit fare structure and level can largely affect passengers’travel behavior and route choices.The commonly used transit fare policies in the present transit network would lead to the unbalanced transit assignment and improper transit resources distribution.In order to distribute transit passenger flow evenly and efficiently,this paper introduces a new distance-based fare pattern with Euclidean distance.A bi-level programming model is developed for determining the optimal distance-based fare pattern,with the path-based stochastic transit assignment(STA)problem with elastic demand being proposed at the lower level.The upper-level intends to address a principal-agent game between transport authorities and transit enterprises pursing maximization of social welfare and financial interest,respectively.A genetic algorithm(GA)is implemented to solve the bi-level model,which is verified by a numerical example to illustrate that the proposed nonlinear distance-based fare pattern presents a better financial performance and distribution effect than other fare structures.

DISTANCE-BASED UPDATE STRATEGY IN LDCQ

A new update strategy, distance-based update strategy, is presented in Location Dependent Continuous Query (LDCQ) under error limitation. There are different possibilities to intersect when the distances between moving objects and the querying boundary are different.Therefore, moving objects have different influences to the query result. We set different deviation limits for different moving objects according to distances. A great number of unnecessary updates are reduced and the payload of the system is relieved.

A Study of Detection of Outliers for Working and Non-Working Days Air Quality in Kolkata, India: A Case Study

A variety of factors affect air quality, making it a difficult issue. The level of clean air in a certain area is referred to as air quality. It is challenging for conventional approaches to correctly discover aberrant values or outliers due to the significant fluctuation of this sort of data, which is influenced by Climate change and the environment. With accelerating industrial expansion and rising population density in Kolkata City, air pollution is continuously rising. This study involves two phases, in the first phase imputation of missing values and second detection of outliers using Statistical Process Control (SPC), and Functional Data Analysis (FDA), studies to achieve the efficacy of the outlier identification methodology proposed with working days and Nonworking days of the variables NO2, SO2, and O3, which were used for a year in a row in Kolkata, India. The results show how the functional data approach outshines traditional outlier detection methods. The outcomes show that functional data analysis vibrates more than the other two approaches after imputation, and the suggested outlier detector is absolutely appropriate for the precise detection of outliers in highly variable data.

CLOF Based Outlier Detection Algorithm of Temperature Data for Ethylene Cracking Furnace

The flue temperature is one of the important indicators to characterize the combustion state of an ethylene cracker furnace,the outliers of temperature data can lead to the false alarm.Conventional outlier detection algorithms such as the Isolation Forest algorithm and 3-sigma principle cannot detect the outliers accurately.In order to improve the detection accuracy and reduce the computational complexity,an outlier detection algorithm for flue temperature data based on the CLOF(Clipping Local Outlier Factor,CLOF)algorithm is proposed.The algorithm preprocesses the normalized data using the cluster pruning algorithm,and realizes the high accuracy and high efficiency outlier detection in the outliers candidate set.Using the flue temperature data of an ethylene cracking furnace in a petrochemical plant,the main parameters of the CLOF algorithm are selected according to the experimental results,and the outlier detection effect of the Isolation Forest algorithm,the 3-sigma principle,the conventional LOF algorithm and the CLOF algorithm are compared and analyzed.The results show that the appropriate clipping coefficient in the CLOF algorithm can significantly improve the detection efficiency and detection accuracy.Compared with the outlier detection results of the Isolation Forest algorithm and 3-sigma principle,the accuracy of the CLOF detection results is increased,and the amount of data calculation is significantly reduced.

A Novel Outlier Detection with Feature Selection Enabled Streaming Data Classification

Due to the advancements in information technologies,massive quantity of data is being produced by social media,smartphones,and sensor devices.The investigation of data stream by the use of machine learning(ML)approaches to address regression,prediction,and classification problems have received consid-erable interest.At the same time,the detection of anomalies or outliers and feature selection(FS)processes becomes important.This study develops an outlier detec-tion with feature selection technique for streaming data classification,named ODFST-SDC technique.Initially,streaming data is pre-processed in two ways namely categorical encoding and null value removal.In addition,Local Correla-tion Integral(LOCI)is used which is significant in the detection and removal of outliers.Besides,red deer algorithm(RDA)based FS approach is employed to derive an optimal subset of features.Finally,kernel extreme learning machine(KELM)classifier is used for streaming data classification.The design of LOCI based outlier detection and RDA based FS shows the novelty of the work.In order to assess the classification outcomes of the ODFST-SDC technique,a series of simulations were performed using three benchmark datasets.The experimental results reported the promising outcomes of the ODFST-SDC technique over the recent approaches.

Outliers rejection in similar image matching

Background Image matching is crucial in numerous computer vision tasks such as 3D reconstruction and simultaneous visual localization and mapping.The accuracy of the matching significantly impacted subsequent studies.Because of their local similarity,when image pairs contain comparable patterns but feature pairs are positioned differently,incorrect recognition can occur as global motion consistency is disregarded.Methods This study proposes an image-matching filtering algorithm based on global motion consistency.It can be used as a subsequent matching filter for the initial matching results generated by other matching algorithms based on the principle of motion smoothness.A particular matching algorithm can first be used to perform the initial matching;then,the rotation and movement information of the global feature vectors are combined to effectively identify outlier matches.The principle is that if the matching result is accurate,the feature vectors formed by any matched point should have similar rotation angles and moving distances.Thus,global motion direction and global motion distance consistencies were used to reject outliers caused by similar patterns in different locations.Results Four datasets were used to test the effectiveness of the proposed method.Three datasets with similar patterns in different locations were used to test the results for similar images that could easily be incorrectly matched by other algorithms,and one commonly used dataset was used to test the results for the general image-matching problem.The experimental results suggest that the proposed method is more accurate than other state-of-the-art algorithms in identifying mismatches in the initial matching set.Conclusions The proposed outlier rejection matching method can significantly improve the matching accuracy for similar images with locally similar feature pairs in different locations and can provide more accurate matching results for subsequent computer vision tasks.

Outlier Detection of Air Quality for Two Indian Urban Cities Using Functional Data Analysis

Human living would be impossible without air quality. Consistent advancements in practically every aspect of contemporary human life have harmed air quality. Everyday industrial, transportation, and home activities turn up dangerous contaminants in our surroundings. This study investigated two years’ worth of air quality and outlier detection data from two Indian cities. Studies on air pollution have used numerous types of methodologies, with various gases being seen as a vector whose components include gas concentration values for each observation per-formed. We use curves to represent the monthly average of daily gas emissions in our technique. The approach, which is based on functional depth, was used to find outliers in the city of Delhi and Kolkata’s gas emissions, and the outcomes were compared to those from the traditional method. In the evaluation and comparison of these models’ performances, the functional approach model studied well.

基于期望核密度离群因子的离群点检测算法

针对基于密度的离群点检测方法在不同分布的数据集上检测精度低的问题,提出了一种基于期望核密度离群因子的离群点检测算法。首先,引入k近邻和反向k近邻扩展邻域空间(ENS)代替传统的k邻域范围,更加全面地考虑数据对象的邻域信息;其次,在传统核密度估计(KDE)方法的基础上引入多元高斯函数,在扩展邻域空间内估计数据对象的密度,同时借鉴自适应核带宽的思想,更好地适应不同数据集的数据分布;然后,给出期望距离的概念,进一步区分局部离群点和位于低密度区域的正常点;最后,定义了期望核密度离群因子刻画数据对象离群程度。在人工数据集和真实数据集上对所提算法进行实验验证,并与部分传统算法进行对比,验证了所提算法的有效性。

基于贡献度和数据有效性检验的共识机制

将区块链技术引入到分布式数据维护系统,旨在解决基于传统中心化数据库的分布式系统存在的数据维护不透明、数据易被篡改、历史记录不可追溯等问题,提出一种基于贡献度和数据有效性检验的共识机制.该算法提出一种贡献度优先的随机可验证领导者选举机制,保证记账权分配的随机性及可验证性.进一步引入密度峰值算法对交易数据有效性进行校验,对打包区块的正确性达成共识.最后将所提出的共识机制应用于梅花鹿分布式养殖场场景,结果验证了密度峰值算法在交易数据有效性检测任务中的准确性和高效性.出块时延分析和安全性分析表明,所提出的共识机制能够满足数据有效性验证的实时性需求,能耗较小,具有很强的灾备能力.

基于5G移动通信技术在医疗设备管理中的研究

目的 为解决现有医疗设备通信技术滞后、设备管理能力欠佳等问题,提出一种基于5G移动通信技术在医疗设备管理中的方法,以提高医疗设备之间的数据信息通信能力。方法 基于5G通信技术,融合安全传输层协议,对传输数据进行加密;结合WebRTC技术,实现远程视频通话和会诊;使用离群检测算法对数据进行处理,提高通信数据的信息检索能力;加入网络波动检测模块,降低在远程访问中网络波动带来的影响。结果 通过记录某医疗机构某年1—10月的远程视频次数及网络波动次数和触发网络加速模块的次数可知,本系统将目标特征误差保持在2%以下,在系统运行过程中非常稳定且适用。结论 本系统大大减轻了系统主机的压力,具有较高的工作效率及较低的误差率和波动率,可达到远程医用和远程急救的要求。

基于映射距离比离群因子的离群点检测算法

针对基于邻近性的离群点检测方法需要花费大量时间过滤正常点,并且在检测全局离群点时难以检测出局部离群点的问题,提出一种基于映射距离比离群因子离群点检测(MDROF)算法。首先,为了减少正常点在检测过程中的时间消耗,给出了差异相似度的概念,通过定义差异相似度剪枝因子过滤掉数据集中的大部分正常点。其次,定义映射k距离,通过映射距离与可达距离的比值刻画数据对象的局部离群程度,通过可达密度刻画数据对象的全局离群程度。最后,结合数据对象相互近邻点的平均排位定义映射距离比离群因子来检测离群点。在人工数据集以及真实数据集上分别对该算法与其他经典的离群点检测算法在精确率、AUC值和离群点发现曲线上进行实验对比分析。实验结果证明MDROF算法在离群点检测的准确性和稳定性上明显优于对比算法。

基于自适应高斯渐进滤波的工程车GNSS/INS紧组合定位

研究了量测野值影响下的工程车GNSS/INS紧组合定位问题,提出了一种基于自适应高斯渐进滤波的车辆定位方法。首先,为降低量测野值对滤波器的破坏风险,利用假设检验方法对量测野值进行检测和剔除;其次,对于野值漏检测引起的定位性能下降的问题,设计了自适应的高斯渐进滤波方法来补偿量测的不确定性;特别地,利用线性化误差与系统估计误差的变化关系,对渐进量测更新方式进行了改进,从而实现对线性化误差的间接补偿。最后,通过工程车GNSS/INS紧组合定位实验进行结果分析,验证了所提方法的可靠性和优越性。

启发式k-means聚类算法的改进研究

启发式k-means聚类算法通过在k-means第一次迭代后查看附近的集群来预测每个数据点可能会被划分到的集群子集,有效地加快了算法的运行速度。但由于启发式算法存在随机选择初始聚类中心以及无法有效识别数据集中离群点的缺陷,导致聚类结果的误差平方和较大并且轮廓系数偏小。针对这一问题,提出了CHk-means算法,该算法引入仔细播种方法,克服了启发式k-means算法随机选择初始聚类中心带来的局部最优解问题;该算法引入局部异常因子LOF算法对离群点进行检测,降低了离群点数据对聚类结果的影响。在多个数据集上对3种算法进行对比试验,结果表明CHk-means算法可有效降低聚类结果的误差平方和,增强聚类的轮廓系数,使聚类质量得到明显改善。

颠覆性技术识别与扩散趋势预测:概念模型与实证分析

发现并判断技术颠覆性潜力和扩散趋势,能够为国家和政府科技资源分配与未来产业的超前布局提供精准决策依据。首先,本文构建了颠覆性技术识别与扩散趋势预测的概念模型;其次,依据此模型以3D打印领域为例,从离群性和影响力维度识别颠覆性专利,提取颠覆性技术;最后,基于识别的颠覆性专利的施引专利,将自动标签和战略坐标应用于技术主题扩散路径绘制中,提出一种新的多位态自动标签技术主题扩散趋势预测方法,用于揭示核心、边缘、成熟、新兴等位态主题之间的动态扩散关系。研究发现,离群专利与颠覆性技术之间存在共生、匹配和关联的内在逻辑关系,从离群专利视角识别颠覆性技术具有可行性;1955—2017年,3D打印领域的颠覆性技术主要分布在高端装备制造、生物医药和材料3大方向,突出的技术领域是运输、发动机/泵/涡轮机、生物材料分析、半导体、环境技术;多位态自动标签技术主题扩散趋势预测结果显示,生物医疗3D打印技术主题未来发展潜力巨大。

高校高价值专利技术机会识别研究——以“生成式人工智能”领域为例

提出一种高校高价值专利技术机会识别方法,使用主题建模、突变级数法、机器学习与离群值检测算法,在评估出高校高价值专利的基础上,进一步识别出具有潜在技术机会的技术主题与专利技术。以“生成式人工智能”领域为例进行实证,研究结果表明:“生成式人工智能”领域的潜在技术主题集中在深度学习、神经网络与机器学习等前沿领域,AI影像、AI诊疗等技术为该领域的潜在技术机会,且上述技术均有国家相关政策大力支撑。本研究方法突破了单一技术机会识别方法识别结果针对性不强、识别专利价值不大、识别结果形式较为单一等核心问题,相关识别结果可以为高校技术转移、技术研发与技术创新提供决策支撑。

基于离群点检测和自适应参数的三支DBSCAN算法

针对经典的DBSCAN算法存在难以确定全局最优参数和误判离群点的问题,该算法首先从选择最优参数角度出发,通过数据集的分布特征生成Eps和MinPts列表,将两个列表中的参数进行全组合操作,把不同的参数组合依次进行聚类,从而寻找准确率最高点对应的参数。最后从离群点角度出发,将三支决策思想与离群点检测LOF算法进行结合。该算法与多种聚类算法进行效果对比分析,结果表明该算法能够全自动化选择全局最优参数,并提高聚类算法的准确性。