面向移动对象连续k近邻查询的双层索引结构

移动对象连续k近邻（CKNN）查询是指给定一个连续移动的对象集合,对于任意一个k近邻查询q,实时计算查询q的k近邻并在查询有效时间内对查询结果进行实时更新.现实生活中,交通出行、社交网络、电子商务等领域许多基于位置的应用服务都涉及移动对象连续k近邻查询这一基础问题.已有研究工作解决连续k近邻查询问题时,大多需要通过多次迭代确定一个包含k近邻的查询范围,而每次迭代需要根据移动对象的位置计算当前查询范围内移动对象的数量,整个迭代过程的计算代价占查询代价的很大部分.为此,提出了一种基于网络索引和混合高斯函数移动对象分布密度的双重索引结构（grid GMM index,GGI）,并设计了移动对象连续k近邻增量查询算法（incremental search for continuous k nearest neighbors,IS-CKNN）.GGI索引结构的底层采用网格索引对海量移动对象进行维护,上层构建混合高斯模型模拟移动对象在二维空间中的分布.对于给定的k近邻查询q,IS-CKNN算法能够基于混合高斯模型直接确定一个包含q的k近邻的查询区域,减少了已有算法求解该区域的多次迭代过程;当移动对象和查询q位置发生变化时,进一步提出一种高效的增量查询策略,能够最大限度地利用已有查询结果减少当前查询的计算量.最后,在滴滴成都网约车数据集以及两个模拟数据集上进行大量实验,充分验证了算法的性能.     

基于受限移动对象当前及将来时刻检索的时空索引结构研究

在给定的空间及时间范围内,如何构建高效的时空索引结构,以实现对移动对象快速有效的检索,是实现定位服务、智能交通、数字化战争等诸多应用中所迫切需要解决的问题.本文依据移动对象的运动特点,提出了一种面向当前及将来时刻快速更新及有效检索的索引结构—PQR树.PQR树是综合PMRQuad树和R*树的结构,首先依据道路分布用PMRQuad树将移动对象的索引空间实行粗略的层分割,将所有快速移动对象与道路相关联.然后用R*树索引分布在各个子空间块内的类静止对象.实验结果表明PQR树具有良好的更新和查询性能.     

保护位置隐私和查询内容隐私的路网K近邻查询方法

位置隐私和查询内容隐私是LBS兴趣点(point of interest,简称POI)查询服务中需要保护的两个重要内容,同时,在路网连续查询过程中,位置频繁变化会给LBS服务器带来巨大的查询处理负担,如何在保护用户隐私的同时,高效地获取精确查询结果,是目前研究的难题.以私有信息检索中除用户自身外其他实体均不可信的思想为基本假设,基于Paillier密码系统的同态特性,提出了无需用户提供真实位置及查询内容的K近邻兴趣点查询方法,实现了对用户位置、查询内容隐私的保护及兴趣点的精确检索;同时,以路网顶点为生成元组织兴趣点分布信息,进一步解决了高强度密码方案在路网连续查询中因用户位置变化频繁导致的实用效率低的问题,减少了用户的查询次数,并能确保查询结果的准确性.最后从准确性、安全性及查询效率方面对本方法进行了分析,并通过仿真实验验证了理论分析结果的正确性.     

Fast Nearest-Neighbor Query Processing in Moving-Object Databases

A desirable feature in spatio-temporal databases is the ability to answer future queries, based on the current data characteristics (reference position and velocity vector). Given a moving query and a set of moving objects, a future query asks for the set of objects that satisfy the query in a given time interval. The difficulty in such a case is that both the query and the data objects change positions continuously, and therefore we can not rely on a given fixed reference position to determine the answer. Existing techniques are either based on sampling, or on repetitive application of time-parameterized queries in order to provide the answer. In this paper we develop an efficient method in order to process nearest-neighbor queries in moving-object databases. The basic advantage of the proposed approach is that only one query is issued per time interval. The time-parameterized R-tree structure is used to index the moving objects. An extensive performance evaluation, based on CPU and I/O time, shows that significant improvements are achieved compared to existing techniques.     

移动对象历史轨迹的连续最近邻查询算法

提出一种直接在原始时空坐标系下对一维移动对象的历史轨迹进行连续最近邻查询的算法,给出相关的定义与定理,通过分析轨迹之间的交点及轨迹线段单调性等特征,利用以前所求得的最近邻结果信息进行连续最近邻的查询判断。实验结果与分析表明,该算法具有较好的性能。     

移动对象的连续最近邻查询算法

介绍了一种索引结构———TPR树和静态环境中基本的最近邻查询算法,并提出了影响时间这一概念,将其运用到最近邻查询算法中,可以完成移动对象的连续最近邻查询。     

Oporto: A Realistic Scenario Generator for Moving Objects

The spatio-temporal database research community has just started to investigate benchmarking issues. On one hand we would rather have a benchmark that is representative of real world applications, in order to verify the expressiveness of proposed models. On the other hand, we would like a benchmark that offers a sizeable workload of data and query sets, which could obviously stress the strengths and weaknesses of a broad range of data access methods. This paper offers a framework for a spatio-temporal data sets generator, a first step towards a full benchmark for the large real world application field of smoothly moving objects with few or no restrictions in motion. The driving application is the modeling of fishing ships where the ships go in the direction of the most attractive shoals of fish while trying to avoid storm areas. Shoals are themselves attracted by plankton areas. Ships are moving points; plankton or storm areas are regions with fixed center but moving shape; and shoals are moving regions. The specification is written in such a way that the users can easily adjust generation model parameters.     

路网中查询点速度不确定的连续k近邻查询方法

提出一种路网中查询点速度不确定的连续k近邻查询方法.查询点在起始位置向服务器提出查询请求,得到k近邻的候选集.随着查询点的移动,利用有效候选集计算当前的k近邻,而不必再向服务器请求,从而减少了服务器计算代价.当候选集部分失效时,由服务器返回候选集中失效的兴趣点的当前信息,使候选集有效.当候选集完全失效时,由查询点重新向服务器提出查询请求,得到新的候选集.并提出一种计算候选集的优化方法,降低了查询代价.最后,通过实验验证了所提算法的有效性.     

一种移动对象增量组最近邻居查询方法

基于移动对象增量最近邻居查询(INCNN),提出一种移动对象增量组最近邻居查询方法(INCGNN)。该方法以较小的代价求出每次更新时刻的组最近邻居。针对组最近邻居查询的特点,给出了k+1组最近邻居查询改进算法,椭圆剪枝和延迟更新3种优化查询的策略。实验结果表明,INCGNN明显优于REEGNN,3种优化策略能较好地提高查询性能。     

Efficient k-Nearest-Neighbor Search Algorithms for Historical Moving Object Trajectories

k Nearest Neighbor (kNN) search is one of the most important operations in spatial and spatio-temporal databases. Although it has received considerable attention in the database literature, there is little prior work on kNN retrieval for moving object trajectories. Motivated by this observation, this paper studies the problem of efficiently processing kNN (k≥1) search on R-tree-like structures storing historical information about moving object trajectories. Two algorithms are developed based on best-first traversal paradigm, called BFPkNN and BFTkNN, which handle the kNN retrieval with respect to the static query point and the moving query trajectory, respectively. Both algorithms minimize the number of node access, that is, they perform a single access only to those qualifying nodes that may contain the final result. Aiming at saving main-memory consumption and reducing CPU cost further, several effective pruning heuristics are also presented. Extensive experiments with synthetic and real datasets confirm that the proposed algorithms in this paper outperform their competitors significantly in both efficiency and scalability.     

Indexing the Trajectories of Moving Objects in Networks*

The management of moving objects has been intensively studied in recent years. A wide and increasing range of database applications has to deal with spatial objects whose position changes continuously over time, called moving objects. The main interest of these applications is to efficiently store and query the positions of these continuously moving objects. To achieve this goal, index structures are required. The main proposals of index structures for moving objects deal with unconstrained 2-dimensional movement. Constrained movement is a special and a very important case of object movement. For example, cars move in roads and trains in railroads. In this paper we propose a new index structure for moving objects on networks, the MON-Tree. We describe two network models that can be indexed by the MON-Tree. The first model is edge oriented, i.e., the network consists of nodes and edges and there is a polyline associated to each edge. The second one is more suitable for transportation networks and is route oriented, i.e., the network consists of routes and junctions. In this model, a polyline also serves as a representation of the routes. We propose the index in terms of the basic algorithms for insertion and querying. We test our proposal in an extensive experimental evaluation with generated data sets using as underlying networks the roads of Germany. In our tests, the MON-Tree shows good scalabiliy and outperforms the competing index structures in updating (index creation) as well as in querying.*This work was partially supported by a grant Gu 293/8–l from the Deutsche Forschungsgemeinschaft (DFG), project Datenbanken für bewegte Objekte (Databases for Moving Objects).     

一种障碍空间数据库中的连续反k近邻查询方法

随着智能移动设备和无线定位技术的飞速发展,使用基于位置服务应用的用户越来越多.特别地,不同于传统的针对固定位置的快照查询,移动的用户往往基于移动轨迹发出连续的查询.在真实和虚拟的空间环境中,障碍物的影响都是广泛存在的,障碍空间内的查询处理技术得到了越来越多的关注,其中,障碍空间内的连续反k近邻查询处理有着重要的应用.对障碍空间中的连续反k近邻查询问题进行了定义和系统的研究,通过定义控制点和分割点,提出了针对该问题的处理框架.进一步地,提出了一系列的过滤和求精算法,包括剪枝数据集、获取障碍物、剪枝和计算控制点和更新结果集等处理策略.基于多种数据集对所提出的算法进行了实验评估.与针对每个数据点进行k 近邻计算的基本方法相比,这些方法可以大幅度提高查询处理的CPU 和I/O 效率.     

An Enhanced Technique for k-Nearest Neighbor Queries with Non-Spatial Selection Predicates

In multimedia databases, k-nearest neighbor queries are popular and frequently contain non-spatial predicates. Among the available techniques for such queries, the incremental nearest neighbor algorithm proposed by Hjaltason and Samet is known as the most useful algorithm [16]. The reason is that if k > k neighbors are needed, it can provide the next neighbor for the upper operator without restarting the query from scratch. However, the R-tree in their algorithm has no facility capable of partially pruning tuple candidates that will turn out not to satisfy the remaining predicates, leading their algorithm to inefficiency. In this paper, we propose an RS-tree-based incremental nearest neighbor algorithm complementary to their algorithm. The RS-tree used in our algorithm is a hybrid of the R-tree and the S-tree, as its buddy tree, based on the hierarchical signature file. Experimental results show that our RS-tree enhances the performance of Hjaltason and Samet's algorithm.     

时空数据库中多个最近邻对象的查询算法

在时空数据库中,最近邻查询用于对某个查询对象,在被查询对象中找出离它最近的一个或多个对象。该文在TPR树这一时空索引的基础上,提出了一种高效的最近邻查询算法,能够支持移动对象的多个最近邻对象的查询,并在性能上也有所提高。     

Indexing Fast Moving Objects for kNN Queries Based on Nearest Landmarks

With the rapid advancements in positioning technologies such as the Global Positioning System (GPS) and wireless communications, the tracking of continuously moving objects has become more convenient. However, this development poses new challenges to database technology since maintaining up-to-date information regarding the location of moving objects incurs an enormous amount of updates. Existing indexes can no longer keep up with the high update rate while providing speedy retrieval at the same time. This study aims to improve k nearest neighbor (kNN) query performance while reducing update costs. Our approach is based on an important observation that queries usually occur around certain places or spatial landmarks of interest, called reference points. We propose the Reference-Point-based tree (RP-tree), which is a two-layer index structure that indexes moving objects according to reference points. Experimental results show that the RP-tree achieves significant improvement over the TPR-tree.

基于K近邻和优化分配策略的密度峰值聚类算法

密度峰值聚类(density peak clustering, DPC)是一种简单有效的聚类分析方法.但在实际应用中,对于簇间密度差别大或者簇中存在多密度峰的数据集,DPC很难选择正确的簇中心;同时,DPC中点的分配方法存在多米诺骨牌效应.针对这些问题,提出一种基于K近邻(K-nearest neighbors,KNN)和优化分配策略的密度峰值聚类算法.首先,基于KNN、点的局部密度和边界点确定候选簇中心;定义路径距离以反映候选簇中心之间的相似度,基于路径距离提出密度因子和距离因子来量化候选簇中心作为簇中心的可能性,确定簇中心.然后,为了提升点的分配的准确性,依据共享近邻、高密度最近邻、密度差值和KNN之间距离构建相似度,并给出邻域、相似集和相似域等概念,以协助点的分配;根据相似域和边界点确定初始聚类结果,并基于簇中心获得中间聚类结果.最后,依据中间聚类结果和相似集,从簇中心到簇边界将簇划分为多层,分别设计点的分配策略;对于具体层次中的点,基于相似域和积极域提出积极值以确定点的分配顺序,将点分配给其积极域中占主导地位的簇,获得最终聚类结果.在11个合成数据集和27个真实数据集上进行仿真...     

道路网络中移动对象的索引研究

为了能有效地实现网络中移动对象的过去、当前和将来轨迹的查询,提出了一种L2R索引,它由两层R树和一个链表结构组成。两层R树用以索引道路网络和移动对象过去的运动,对象当前的位置和将来的预测轨迹信息保存在链表中。L2R索引不仅可以支持网络中的移动对象的轨迹查询,尤其是可方便的在纵向链表中查询在同条路线上的所有对象。在此索引基础上文中实施了对移动对象的范围查询和点查询,最后通过实验表明L2R结构的索引和查询性能均要优越于TPR树。     

Classification of weld flaws with imbalanced class data

This paper presents research results of our investigation of the imbalanced data problem in the classification of different types of weld flaws, a multi-class classification problem. The one-against-all scheme is adopted to carry out multi-class classification and three algorithms including minimum distance, nearest neighbors, and fuzzy nearest neighbors are employed as the classifiers. The effectiveness of 22 data preprocessing methods for dealing with imbalanced data is evaluated in terms of eight evaluation criteria to determine whether any method would emerge to dominate the others. The test results indicate that: (1) nearest neighbor classifiers outperform the minimum distance classifier; (2) some data preprocessing methods do not improve any criterion and they vary from one classifier to another; (3) the combination of using the AHC_KM data preprocessing method with the 1-NN classifier is the best because they together produce the best performance in six of eight evaluation criteria; and (4) the most difficult weld flaw type to recognize is crack.     

A privacy preserving technique for distance-based classification with worst case privacy guarantees

There has been relatively little work on privacy preserving techniques for distance based mining. The most widely used ones are additive perturbation methods and orthogonal transform based methods. These methods concentrate on privacy protection in the average case and provide no worst case privacy guarantee. However, the lack of privacy guarantee makes it difficult to use these techniques in practice, and causes possible privacy breach under certain attacking methods. This paper proposes a novel privacy protection method for distance based mining algorithms that gives worst case privacy guarantees and protects the data against correlation-based and transform-based attacks. This method has the following three novel aspects. First, this method uses a framework to provide theoretical bound of privacy breach in the worst case. This framework provides easy to check conditions that one can determine whether a method provides worst case guarantee. A quick examination shows that special types of noise such as Laplace noise provide worst case guarantee, while most existing methods such as adding normal or uniform noise, as well as random projection method do not provide worst case guarantee. Second, the proposed method combines the favorable features of additive perturbation and orthogonal transform methods. It uses principal component analysis to decorrelate the data and thus guards against attacks based on data correlations. It then adds Laplace noise to guard against attacks that can recover the PCA transform. Third, the proposed method improves accuracy of one of the popular distance-based classification algorithms: K-nearest neighbor classification, by taking into account the degree of distance distortion introduced by sanitization. Extensive experiments demonstrate the effectiveness of the proposed method.