一种基于密度的快速聚类算法

聚类是数据挖掘领域中的一个重要研究方向,聚类技术在统计数据分析、模式识别、图像处理等领域有广泛应用,迄今为止人们提出了许多用于大规模数据库的聚类算法。基于密度的聚类算法DBSCAN就是一个典型代表。以DBSCAN为基础,提出了一种基于密度的快速聚类算法。新算法以核心对象领域中所有对象的代表对象为种子对象来扩展类,从而减少区域查询次数,降低I/O开销,实现快速聚类,对二维空间数据测试表明：快速算法能够有效地对大规模数据库进行聚类,速度上数倍于已有DBSCAN算法。     

Identifying Objects in Procedural Programs Using Clustering Neural Networks

This paper presents a general approach for the identification of objects in procedural programs. The approach is based on neural architectures that perform an unsupervised learning of clusters. We describe two such neural architectures, explain how to use them in identifying objects in software systems and briefly describe a prototype tool, which implements the clustering algorithms. With the aid of several examples, we explain how our approach can identify abstract data types as well as groups of routines which reference a common set of data. The clustering results are compared to the results of many other object identification techniques. Finally, several case studies were performed on existing programs to evaluate the object identification approach. Results concerning two representative programs and their generated clusters are discussed.     

A Linear Algorithm for Computing the Homography from Conics in Correspondence

This paper presents a study, based on conic correspondences, on the relationship between two perspective images acquired by an uncalibrated camera. We show that for a pair of corresponding conics, the parameters representing the conics satisfy a linear constraint. To be more specific, the parameters that represent a conic in one image are transformed by a five-dimensional projective transformation to the parameters that represent the corresponding conic in another image. We also show that this transformation is expressed as the symmetric component of the tensor product of the transformation based on point/line correspondences and itself. In addition, we present a linear algorithm for uniquely determining the corresponding point-based transformation from a given conic-based transformation up to a scale factor. Accordingly, conic correspondences enable us to easily handle both points and lines in uncalibrated images of a planar object.     

Triggered Updates for Temporal Consistency in Real-Time Databases

A real-time database systemhas temporal consistency constraints in addition to timing constraints.The timing constraints require a transaction to be completedby a specified deadline, and the temporal consistency constraintsrequire that temporal data read by a transaction be up-to-date.If a transaction reads out-of-date data, it will become temporallyinconsistent. A real-time database system consists of differenttypes of temporal data objects, including derived objects. Thevalue of a derived object is computed from a set of other objects,known as the read-set of the derived object. The derived objectmay not always reflect the current state of its read-set; a derivedobject can become out-of-date even if its read-set is up-to-date.Any subsequent transaction reading the derived object will thenbecome temporally inconsistent. In this case, in order to readup-to-date objects, a transaction will have to wait until someother transaction updates the out-of-date object. However, indoing so, the waiting transaction may miss its deadline, particularlyif the update is not periodic but instead arrives randomly. Wepropose to update the outdated objects so that not only is thetemporal consistency improved, but also the number of misseddeadlines does not increase significantly, and as a result thereis an overall improvement in the performance of the system. Wepropose, implement and study a novel approach, to be known astriggered updates, to improve temporal consistency in firm real-timedatabase systems when updates are not periodic. We identify propertiesof triggered updates and explain how they work by giving bothan intuitive and a probabilistic analysis. We present strategiesfor generating triggered updates, discuss their suitability invarious contexts and perform a detailed simulation study to evaluatetheir performance. Results show that it is possible to improvetemporal consistency without degrading the timeliness of real-time database systems to a great deal.     

Primal or dual: which promises faster spatiotemporal search?

The existing predictive spatiotemporal indexes can be classified into two categories, depending on whether they are based on the primal or dual methodology. Although we have gained considerable empirical knowledge about various access methods, currently there is only limited understanding on the theoretical characteristics of the two methodologies. In fact, the experimental results in different papers even contradict each other, regarding the relative superiority of the primal and dual techniques. This paper presents a careful study on the query performance of general primal and dual indexes, and reveals important insight into the behavior of each technique. In particular, we mathematically establish the conditions that determine the superiority of each methodology, and provide rigorous justification for well-known observations that have not been properly explained in the literature. Our analytical findings also resolve the contradiction in the experiments of previous work.     

A grid-based node split algorithm for managing current location data of moving objects

There is rapidly increasing interest in Location Based Service (LBS) which utilizes location data of moving objects. To efficiently manage the huge amounts of location data in LBS, the GALIS (Gracefully Aging Location Information System) architecture, a cluster-based distributed computing architecture, is proposed. The GALIS using the non-uniform 2-level grid algorithm performs load balancing and indexing for nodes. However, the non-uniform 2-level grid algorithm has a problem creating unnecessary nodes when moving objects are crowded in a certain region. Therefore, a new node split algorithm, which is more efficient for various distribution of moving objects, is proposed in this paper. Because the algorithm proposed in this paper considers spatial distribution for the current location of moving objects, it can perform efficient load balancing without creating unnecessary nodes even when moving objects are congested in a certain region. Besides, the various data distribution configuration for moving objects has been experimented by implementing node split simulators and it’s been verified that the proposed algorithm can split nodes more efficiently than the existing algorithm.

Limitations of nonlinear discrete periodic control for disturbance attenuation and robust stabilization

This paper presents a performance analysis of nonlinear periodically time-varying discrete controllers acting upon a linear time-invariant discrete plant. Time-invariant controllers are distinguished from strictly periodically time-varying controllers. For a given nonlinear periodic controller, a time-invariant controller is constructed. Necessary and sufficient conditions are given under which the time-invariant controller gives strictly better control performance than the time-invariant controller from which it was obtained, for the attenuation of l_p exogenous disturbances and the robust stabilization of l_p unstructured perturbations, for all p[1,∞].     

A method of predicting the number of clusters using Rand's statistic

Distributional and asymptotic results on the moment of Rand's C_k statistic were derived by DuBien and Warde [1981. Some distributional results concerning a comparative statistic used in cluster analysis. ASA Proceedings of the Social Statistics Section, 309–313.]. Based on those results, a method to predict the number of clusters is suggested by applying various agglomerative clustering algorithms. In the procedure, the methods using different indexes are examined and compared based on the concept of agreement (or, disagreement) between clusterings generated by different clustering algorithms on the set of data. Our method having practical generality works better than the other methods and assigns statistical meaning to C_k values in determining the number of clusters from the comparison.     

d-Dimensional parametric models for dynamic animation of deformable objects

This paper introduces an accurate, efficient, and unified engine dedicated to dynamic animation of d-dimensional deformable objects. The objects are modelled as d-dimensional manifolds defined as functional combinations of a mesh of 3D control points, weighted by parametric blending functions. This model ensures that, at each time step, the object shape conforms to its manifold definitions. The object motion is deduced from the control points dynamic animation. In fact, control points should be viewed as the degrees of freedom of the continuous object. The chosen dynamic equations (Lagrangian formalism) reflect this generic modelling scheme and yield an exact and computationally efficient linear system.     

雨型对东北黑土区坡耕地土壤侵蚀影响的试验研究

降雨雨型对土壤侵蚀过程有重要的影响,也是当前土壤侵蚀研究的热点之一。通过野外原位人工模拟降雨试验,设计了次降雨中平均降雨强度、总降雨量相同的4种不同降雨雨型(均匀型:降雨强度为60mm/h;峰值型:降雨强度分布为30-90-60mm/h;延迟型:降雨强度分布为60-30-90mm/h;减弱型:降雨强度分布为90-60-30mm/h),研究降雨雨型变化对东北黑土区坡耕地土壤侵蚀的影响。结果表明:各雨型条件下休闲坡耕地径流总量的大小顺序为延迟型>减弱性>均匀型>峰值型,而土壤侵蚀总量的大小顺序为减弱型>均匀型>延迟型>峰值型。延迟型雨型的坡面径流总量是其它雨型的1.02~1.45倍,减弱型降雨的侵蚀总量分别是均匀型、延迟型和峰值型的1.03、1.36和2.68倍。同一雨强在不同降雨雨型中出现的位置不同,其产生的坡面径流量和侵蚀量对坡面总径流量、总侵蚀量的贡献率也不同,分布在降雨雨型起始位置的雨强对坡面径流量的贡献率最小;除30mm/h降雨强度外,分布在降雨雨型起始位置的雨强对坡面侵蚀的贡献率最大。研究还发现,试验设计中减弱型降雨雨型与东北黑土区夏季主要侵蚀雨型特点类似,是造成东北黑土区坡耕地土壤侵蚀相对严重的主要雨型。