多维图结构聚类的社交关系挖掘算法

社交关系的数据挖掘一直是大图数据研究领域中的热门问题。图聚类算法如SCAN（Structural clustering algorithm for networks）虽可迅速地从海量图数据中获得关系紧密的社区结构,但这类社区往往只表示了社交对象的聚集,无法反馈对象间的真实社交关系,如家庭成员、同事、同学等。要获取对象间真实的社交关系,需要更多维度地挖掘现实中社交对象间复杂的交互关系。对象间的交互维度很多,例如：通话、见面、微信、Email等,而传统SCAN等聚类算法仅能够挖掘单维度的交互数据。本文在研究社交对象间的多维社交关系图数据与传统图结构聚类算法的基础上,提出了一种有效的子空间聚类算法SCA（Subspace Cluster Algorithm）,首次对多维度下子空间的图结构聚类进行研究,目的是探索如何通过图数据挖掘发现对象间真实的社交关系。SCA算法遵循自底向上的原则,能够发现社交图数据中所有子空间的聚类集。为了提升SCA的运行速度,我们利用其子空间聚类单调性进行了性能优化,进而提出了剪枝算法SCA+。最后,我们进行了大规模的性能测试实验,以及真实数据的案例研究,其结果验证了算法的效率和效用。     

HD Tree: A novel data structure to support multi-dimensional range query for P2P networks

There are two basic concerns for supporting multi-dimensional range query in P2P overlay networks. The first is to preserve data locality in the process of data space partitioning, and the second is the maintenance of data locality among data ranges with an exponentially expanding and extending rate. The first problem has been well addressed by using recursive decomposition schemes, such as Quad-tree, K-d tree, Z-order, and Hilbert curve. On the other hand, the second problem has been recently identified by our novel data structure: HD Tree. In this paper, we explore how data locality can be easily maintained, and how range query can be efficiently supported in HD Tree. This is done by introducing two basic routing strategies: hierarchical routing and distributed routing. Although hierarchical routing can be applied to any two nodes in the P2P system, it generates high volume traffic toward nodes near the root, and has very limited options to cope with node failure. On the other hand, distributed routing concerns source and destination pairs only at the same depth, but traffic load is bound to some nodes at two neighboring depths, and multiple options can be found to redirect a routing request. Because HD Tree supports multiple routes between any two nodes in the P2P system, routing in HD Tree is very flexible; it can be designed for many purposes, like fault tolerance, or dynamic load balancing. Distributed routing oriented combined routing (DROCR) algorithm is one such routing strategy implemented so far. It is a hybrid algorithm combining advantages from both hierarchical routing and distributed routing. The experimental results show that DROCR algorithm achieves considerable performance gain over the equivalent tree routing at the highest depth examined. For supporting multi-dimensional range query, the experimental results indicate that the exponentially expanding and extending rate have been effectively controlled and minimized by HD Tree overlay structure and DROCR routing.     

基于POCS和范例学习的序列图像超分辨率重建

实现序列图像的超分辨率重建,需要利用同一场景的多幅低分辨率图像之间的相对运动信息．并将它们融合到单幅高分辨率图像中,以有效的去除低分辨率图像中的模糊和噪声。本文提出首先分析序列图像结构、纹理等多维特征的不同特性和作用,利用分解得到的多维特征分别采用凸集投影（POCS）、范例学习等具有针对性的重建方法进行图像放大,在有效-融合多维特征重建图像的基础上,实现序列图像的多维特征超分辨率重建。     

关注分离问题研究综述

数十年以来,关注分离问题一直是软件工程的核心问题之一,关注分离是将软件分解成可管理、可理解的部分并将之组织起来的主要动力。本文综合了现有的关注分离的研完工作,分析了对于关注这一个概念的两类观点,并从问题求解的角度给出了关注以及关注分离的定义。论述了在软件开发周期各个阶段从关注分离角度需要考虑的问题、策略和当前的研完热点。     

Optimal representation of multi-dimensional random fields with a moderate number of samples: Application to stochastic mechanics

A significant amount of problems and applications in stochastic mechanics and engineering involve multi-dimensional random functions. The probabilistic analysis of these problems is usually computationally very expensive if a brute-force Monte Carlo method is used. Thus, a technique for the optimal selection of a moderate number of samples effectively representing the entire space of sample realizations is of paramount importance. Functional Quantization is a novel technique that has been proven to provide optimal approximations of random functions using a predetermined number of representative samples. The methodology is very easy to implement and it has been shown to work effectively for stationary and non-stationary one-dimensional random functions. This paper discusses the application of the Functional Quantization approach to the domain of multi-dimensional random functions and the applicability is demonstrated for the case of a 2D non-Gaussian field and a two-dimensional panel with uncertain Young modulus under plane stress.     

多元相图新画法的研究

本文应用“多维西法几何”的基本原理和“不等边浓度图形”表示法,建立了新的多元系浓度立体模型,分析了空间点的多次平行投影过程,提出了独特的多元系浓度表示法,为研究七元系以上的相图画法与应用提供了新的方法。     

S-scheme g-C3N4/TiO2/CFs heterojunction composites with multi-dimensional through-holes and enhanced visible-light photocatalytic activity

A novel multi-dimensional through-holes structure of g-C₃N₄ with adjustable pore size was prepared by controlling the mass ratio of oxamide (OA, structure guiding agent) to urea during one-step calcination process, and a break-rearrangement mechanism was explored. Then, a series of porous g-C₃N₄/TiO₂ (CT) composites with uniformly deposited TiO₂ nanoparticles were prepared based on the multi-dimensional framework by a facile hydrothermal method. The results show that a new S-scheme heterojunction with multi-dimensional through-channel structure was obtained, which is particularly desired for enhancing the visible-light utilization, reducing the carrier recombination rate and enhancing redox capacity. The CT composite obtained at hydrothermal treatment time of 2 h has a specific surface area of 180.15 m² g^-1, which shows high degradation capability (99.99%) for tetracycline hydrochloride (TC·HCl) under 350 W Xe lamp irradiation for 90 min. In addition, CT nanostructures was in-situ growth on carbon fiber (CFs), the degradation rate constant is 0.1566 min^-1, and 90% of the degradation efficiency can be maintained even after 5 consecutive cycles. It is expected to provide an effective reference for solving the problems of recovery difficulty and low reuse rate of powder photocatalytic materials.     

多维角加速度传感器静态标定方法研究

在线加速度传感器静动态标定方法的基础上，根据质量一转动惯量等效原理，提出多维角加速度传感器的静态标定方法。针对一种一体化三维角加速度传感器结构，分析了利用等效力计算等效力矩的过程，进而详细推导出标定的等效数学模型．并在假定结构尺寸的基础之上，计算出具体的等效惯性力矩和等效惯性力，实现了对多维角加速度传感器的静态标定。     

Contextual multi-dimensional browsing

Browsing efficiency depends to a large extent on the organization of the information being browsed. An emerging method of organizing Web content, called multi-faceted categorization, adopts a flat structure and treats different facets of the browsed content as equal and independent. The advantage of this structure is that it allows users to slice and dice the information space from any of the facets they wish to browse, and in any arbitrary order, thereby facilitating so-called multi-dimensional browsing. However, because of limited human information processing capacities, too many choices and too much browsing freedom tend to disorient users and increase the difficulty of information gathering. A more effective means of organizing information is therefore needed.In the literature, it is well documented that context plays a crucial role in making relevance judgments about categorical attributes. Since browsing always occurs in relation to certain contexts, we propose taking context into account for multi-faceted categorization to help organize different facets of information on the browsing interface. Our experiment results demonstrate that the context-sensitive arrangement of categories helps users find information more efficiently. Moreover, users perceive the context-sensitive interface as easier to use than non-contextual interfaces.