一种高效挖掘高维数据的频繁闭合模式算法

为了克服传统高维数据挖掘频繁闭合模式算法迭代产生子表,引起算法执行时间长和存储开销大等问题,提出了一种高效挖掘高维数据的频繁闭合模式的算法EMHCP. EMHCP算法采用一种新型结构位图表来压缩存储数据,在仅扫描数据库一次后,建立位图转换表.根据位图转换表来构建混合树结构,采用深度优先的方式和有效的剪枝策略高效挖掘出所有的闭合模式.从而有效地缩小了搜索空间,加快了处理速度.通过在生物数据库应用的实验结果表明, EMHCP算法比已有的CARPENTER和TD-close等算法更为有效.     

Scenario Reduction Algorithms in Stochastic Programming

We consider convex stochastic programs with an (approximate) initial probability distribution P having finite support supp P, i.e., finitely many scenarios. The behaviour of such stochastic programs is stable with respect to perturbations of P measured in terms of a Fortet-Mourier probability metric. The problem of optimal scenario reduction consists in determining a probability measure that is supported by a subset of supp P of prescribed cardinality and is closest to P in terms of such a probability metric. Two new versions of forward and backward type algorithms are presented for computing such optimally reduced probability measures approximately. Compared to earlier versions, the computational performance (accuracy, running time) of the new algorithms has been improved considerably. Numerical experience is reported for different instances of scenario trees with computable optimal lower bounds. The test examples also include a ternary scenario tree representing the weekly electrical load process in a power management model.     

贵阳二环林带造林栽培技术研究

本文从二环林带的立地条件、栽培方法,栽植树种等方面研究了二环林带的栽培技术,可为以后其它林带的建设提供参考资料.     

基于ULS、 TLS和超声测高仪的天然次生林中不同林冠层树高估测

【目的】应用不同数据源分析不同林冠层中探测提取树高的异同,探索适用于中国北方天然次生林树高估测的方法。【方法】以东北林业大学帽儿山实验林场中林施业区0.25 hm²样地为研究区域,基于无人机激光雷达(unmanned aerial vehicle laser scanning, ULS)、地基激光雷达(terrestrial laser scanning,TLS)和Vertex IV超声测高仪实测单木树高,根据冠层高度分布(canopy height distribution, CHD)对林冠层进行分层,对不同林冠层(上层和下层)、不同树木类型(针叶树和阔叶树)探测提取的树高进行对比与分析。【结果】由CHD计算得到的冠层分层阈值为8.5 m。树高的离群值大多产生在林冠上层,阔叶树比针叶树更容易产生离群值,ULS比TLS更容易产生离群值。在林冠上层,ULS比TLS估测树高的相对均方根误差(rRMSE)低2.56%,ULS提取针叶树树高的rRMSE比阔叶树低2.68%;在林冠下层,ULS仅能探测到少量树木,ULS比TLS探测提取树高的 rRMSE高6.31%,TLS提取针叶树树高的rRMSE比阔叶树低1.16%。【结论】针叶树的树高估测精度普遍高于阔叶树;当TLS和ULS均能对单木进行完全扫描时,具有准确提取树高的潜力;树高离群值多由冠型不规则或相互交叉的阔叶树产生,而大部分针叶树,由于具有规则的冠型,所以产生的离群值较少;基于CHD对林冠层进行划分能够较好地反映不同数据源估测树高的适用范围,具有一定的推广意义。     

基于地面激光点云数据的单木三维重建方法

目的 使用地面三维点云数据,提出一种单株树木三维网格模型重建方法,为精准获取测树因子提供技术支撑。 方法 对获取的点云进行预处理,使用k-d树构建近邻关系图,用Dijkstra算法求算出子图的根。检测出有效路径后,使用探测半径计算关键路径。计算树枝骨架,然后对初始骨架进行Bezier曲线半径平滑,得到平滑的骨架,再将骨架连接,使用半径平滑和圆柱拟合减少点云密度小造成的拟合不足的情况,能够最大限度保留树枝的细节。结果 使用3株落叶松点云数据构建了树枝树干表面网格模型,重建了树木三维结构。将树干、树枝的三维网格模型与点云匹配后,效果较好;所构建的模型能够进行细小枝条的重建,而不是模拟细枝,通过观察重建结果,一级枝的重建效果非常好,大的二级枝也能得到很好的展示;整套算法计算快速,计算时间与枝条的复杂程度、连接关系有关。结论 基于关键路径探测的方法能够很好地构建树木的三维网格模型,可以用于单株树木测树因子的精确提取。     

Set to set broadcasting in communication networks

Suppose G = (V,E) is a graph whose vertices represent people and edges represent telephone lines between pairs of people. Each person knows a unique message and is ignorant of the messages of other people at the beginning. These messages are then spread by telephone calls. In each call, two people exchange all information they have so far in exactly one unit of time. Suppose A and B are two nonempty subsets of V. The main purpose of this paper is to study the minimum number b(A,B,G) of telephone calls by which A broadcasts to B; and the minimum time t(A,B,G) such that A broadcasts to B. In particular, we give an exact formula for b(A,B,K_n) and linear-time algorithms for computing b(A,B,T) and t(A,B,T) of a tree T.     

Phase diagram of the Ising Model on a Cayley tree in the presence of competing interactions and magnetic field

We study the phase diagram for the Ising Model on a Cayley tree with competing nearest-neighbor interactionsJ ₁ and next-nearest-neighbor interactionsJ ₂ andJ ₃ in the presence of an external magnetic field. To perform this study, an iterative scheme similar to that appearing in real space renormalization group frameworks is established; it recovers, as particular cases, previous works by Vannimenus and by Inawashiroet al. At vanishing temperature, the phase diagram is fully determined, for all values and signs ofJ ₂/J ₁ andJ ₃/J ₂; in particular, we verify that values ofJ ₃/J ₂ high enough favor the paramagnetic phase. At finite temperatures, several interesting features (evolution of reentrances, separation of the modulated region into two disconnected pieces, etc.) are exhibited for typical values ofJ ₂/J ₁ andJ ₃/J ₂.Partially supported by the Brazilian Agencies CNPq and FINEP.     

音乐推荐系统中的用户冷启动问题

利用新用户使用音乐时的上下文环境信息来解决冷启动问题.当用户播放音乐时,提取用户当时场景的时间、天气、状态等信息,利用这些信息构建决策树分类模型.当新用户使用音乐推荐系统时,提取当前情景的上下文信息,根据决策树分支和上下文信息判断新用户的兴趣,向新用户推荐音乐.实验结果表明,当采用上下文相关音乐推荐算法,推荐数目不超过6项时,新用户对推荐结果的满意度超过TopN算法.     

Kp,q-factorization of complete bipartite graphs

Let K_(m,n) be a complete bipartite graph with two partite sets having m and nvertices, respectively. A K_(p,q)-factorization of K_(m,n) is a set of edge-disjoint K_(p,q)-factorsof K_(m,n) which partition the set of edges of K_(m,n). When p=i and q is a prime number,Wang, in his paper "On K_(1,k)-factorizations of a complete bipartite graph" (Discrete Math,1994, 126; 359-364), investigated the K_(1,q)-factorization of K_(m,n) and gave a sufficientcondition for such a factorization to exist. In the paper "K_(1,k)-factorizations of completebipartite graphs" (Discrete Math, 2002, 259: 301-306), Du and Wang extended Wang'sresult to the case that q is any positive integer In this paper, we give a sufficient conditionfor K_(m,n) to have a K_(p,q)-factorization. As a special case, it is shown that the Martin's BACconjecture is true when p: q=k: (k+1) for any positive integer k.