FHUQI-Miner: Fast high utility quantitative itemset mining

Applied Intelligence - High utility itemset mining is a popular pattern mining task, which aims at revealing all sets of items that yield a high profit in a transaction database. Although this task...     

到达数据中时空异常聚簇发现

在时空数据中有一类表示用户在某一时间到达某一地点的数据——到达数据,到达数据可以是社交网站的签到数据、轨迹数据中的停留点及公共交通中乘客抵达的位置数据,这些数据的聚簇可以反映用户的聚集行为.基于到达数据,提出一类新的时空数据查询——时空异常聚簇发现.将到达数据进行周期性划分,通过时空聚类算法对一个时间段的数据进行聚类,比较不同时间段内聚簇的差异度,发现具有最大簇异常度的前k个簇.通过该查询发现的时空异常聚簇可以应用于城市安全管理、基于位置的服务和交通调度等方面.定义了异常簇查询模型,提出了针对任意形状聚簇的簇差异度度量,将异常簇查询转化为二分图最大匹配问题,对二分图构建与匹配进行了优化并提出了高效的查询算法.利用真实数据集进行了充分实验,验证了查询结果的实际意义,评估了所提出的各查询算法在不同参数设置下的查询效率.     

Hybrid feature selection by combining filters and wrappers

Feature selection aims at finding the most relevant features of a problem domain. It is very helpful in improving computational speed and prediction accuracy. However, identification of useful features from hundreds or even thousands of related features is a nontrivial task. In this paper, we introduce a hybrid feature selection method which combines two feature selection methods – the filters and the wrappers. Candidate features are first selected from the original feature set via computationally-efficient filters. The candidate feature set is further refined by more accurate wrappers. This hybrid mechanism takes advantage of both the filters and the wrappers. The mechanism is examined by two bioinformatics problems, namely, protein disordered region prediction and gene selection in microarray cancer data. Experimental results show that equal or better prediction accuracy can be achieved with a smaller feature set. These feature subsets can be obtained in a reasonable time period.     

面向对象碰撞检测方法及其在分布式虚拟环境中的应用

本文提出了一种面向对象的碰撞检测方法，分析了这种碰撞检测方法的时间复杂性，讨论了分布式环境中碰撞检测的完全性和唯一性等问题，提出了一种保证分布式虚拟环境中碰撞检测唯一性的方法，最后，讨论了这种方法在分布式虚拟环境ＤＶＥＮＥＴ中的应用。     

快速简单的单触发脉冲扩展器检测纳秒级事件

引言检测快速事件(如光电二极管产生的脉冲或晶体管雪崩事件)要求检测器具有足够的带宽来处理脉冲。LT1711高速比较器可以在4.5ns内切换以响应这类事件(由于其纳秒级的上升时间和传输延迟),但是如果信息需要转发到不能处理持续时间这么短的事件的微处理器或DSP时会怎么样呢?图1     

Evaluation and Improvement of a SVD-Based Empirical Atmospheric Model

T An empirical atmospheric model (EAM) based on the singular value decomposition (SVD) method is evaluated using the composite El Nino/Southern Oscillation (ENSO) patterns of sea surface temperature (SST) and wind anomalies as the target scenario. Two versions of the SVD-based EAM were presented for comparisons. The first version estimates the wind anomalies in response to SST variations based on modes that were calculated from a pair of global wind and SST fields (i.e., conventional EAM or CEAM). The second version utilizes the same model design but is based on modes that were calculated in a region-wise manner by separating the tropical domain from the remaining extratropical regions (i.e., region-wise EAM or REAM). Our study shows that, while CEAM has shown successful model performance over some tropical areas, such as the equatorial eastern Pacific (EEP), the western North Pacific (WNP), and the tropical Indian Ocean (TIO), its performance over the North Pacific (NP) seems poor. When REAM is used to estimate the wind anomalies instead of CEAM, a marked improvement over NP readily emerges. Analyses of coupled modes indicate that such an improvement can be attributed to a much stronger coupled variability captured by the first region-wise SVD mode at higher latitudes compared with that captured by the conventional one. The newly proposed way of constructing the EAM (i.e., REAM) can be very useful in the coupled studies because it gives the model a wider application beyond the commonly accepted tropical domain.     

Electronically coupled porphyrin-arene dyads for dye-sensitized solar cells

An acetylene-linked porphyrin-perylene anhydride and an acetylene-linked porphyrin-naphthalic anhydride have been synthesized; the highly conjugated acetylenic bridge in these porpyrins efficiently mediates electronic interaction between the porphyrin and perylene units to extend the π-conjugation of the porphyrin dye and to cause both broadening and red shifts of both the Soret and Q absorption bands. This condition is a useful feature for efficient dye-sensitized solar cell applications. The optical, electrochemical and photovoltaic properties of the new linked anhydrides show that the HOMO-LUMO gap decreased upon extension of π-conjugation, indicating a strong electronic coupling between the porphyrin and the perylene or naphthalene unit.     

Modified Spherical Wave Functions With Anisotropy Ratio: Application to the Analysis of Scattering by Multilayered Anisotropic Shells

We describe a novel and rigorous vector eigenfunction expansion of electric-type Green's dyadics for radially multi- layered uniaxial anisotropic media in terms of the modified spherical vector wave functions, which can take into account the effects of anisotropy ratio systematically. In each layer, the material constitutions e and epsiv macrmu macr are tensors and distribution of sources is arbitrary. Both the unbounded and scattering dyadic Green's functions (DGFs) for rotationally uniaxial anisotropic media are derived in spherical coordinates (r, thetas, phi). The coefficients of scattering DGFs, based on the coupling recursive algorithm satisfied by the coefficient matrix, are derived and expressed in a compact form. With these DGFs obtained, the electromagnetic fields in each layer are straightforward once the current source is known. A specific model is proposed for the scattering and absorption characteristics of multilayered uniaxial anisotropic spheres, and some novel performance regarding anisotropy effects is revealed.     

基于改进的蛙跳算法的神经网络在语音情感识别中的研究

本文提出了一种改进的混合蛙跳算法,利用混沌运动的遍历性改善初始个体的质量和引入高斯变异,提高了算法的全局搜索能力,同时将改进算法与人工神经网络结合,并把它应用到语音情感识别系统中.依据情感的维度空间模型.分别提取了情感语音的韵律特征与音质特征,研究了谐波噪声比特征随情感类别的变化特性.利用本文所提的蛙跳算法(SFLA)训练随机产生的初始数据,优化神经网络的连接权值,能快速地实现网络的收敛.在实验中比较了BP神经网络、RBF神经网络与改进SFLA神经网络分别用于语音情感以别的识别性能,结果表明基于改进SFLA的神经网络的平均识别率高于BP神经网络9.2个百分点,高于RBF神经网络7.9个百分点.因此本文所提的蛙跳神经网络用于语音情感识别能获得明显的识别性能的提升.     

Segmented Alignment: An Enhanced Model to Align Data Parallel Programs of HPF

In this paper, we propose a new automatic data alignment model called segmented alignment. The conventional data alignment model, such as that used in High-Performance Fortran (HPF), aligns arrays with the whole index domain. The principle of our proposed segmented alignment is to allow alignment relations within delimited index domains. We first provide motivating examples to illustrate how code fragments of HPF with EOSHIFT or CSHIFT operations, or produced by synthesis operations can benefit from our enhanced alignment scheme. Second, we show that this new model can be implemented in HPF-like languages by adding WHEN and IN constructs to them. In addition, we show that the new proposed schemes for WHEN and IN constructs can be emulated using standard HPF syntax. Finally, we address issues related to automatic data alignment for the new proposed model, and present an algorithm to automatically align programs using our segmented alignment scheme. Since the optimal algorithm to do this is NP-hard, a practical heuristic is also given. Our experiments were performed on a DEC Alpha Farm with HPF environments. Our experiments confirm our theory that our proposed alignment scheme can significantly enhance not only the performance of HPF code fragments with EOSHIFT or CSHIFT operations, but also that of codes produced by synthesis operations.