MAXFP-M iner: 利用FP- tree 快速挖掘最大频繁项集

为提高频繁项集的挖掘效率，提出了最大频繁项集树的概念和基于FP-tree的最大频繁项集挖掘算法MAXFP-Miner，首先建立了FP-tree，在此基础上建立最大频繁项集树MAXFP-tree，MAXFP-tree中包含了所有最大频繁项集，缩小了搜索空间，提高了算法的效率，算法分析和实验表明，该算法特别适合于挖掘稠密型及具有长频繁项集的数据集。     

基于矩阵技术的频繁项目集挖掘算法

频繁模式挖掘算法FP-growth算法需递归地生成大量的条件FP-树,且耗费大量存储空间和时间。为此,采用矩阵技术统计约束子树中的频繁项集和频繁项集的支持度,以进行数据挖掘。实验结果表明,该频繁模式挖掘算法是有效的,具有较高的时间效率及空间 效率。     

关联规则挖掘的矩阵算法

关联挖掘作法中的Apriori算法提供了一种根据查找频繁项集来发现数据集中的关联规则的方法，这种算法思路简单易于实现；但在由低次频繁项集生成高次频繁项集时需反复查找数据库，在效率上存在一定的欠缺，在寻找高次频繁项集时尤为明显，文章提出了一种新的关联规则挖掘算法：矩阵算法。同Apriori算法相比较，该算法能直接查找高次频繁项集，可以有效地屏蔽Aptiori算法性能瓶颈试验结果表明，当频繁项级较高时该算法比Apriori具有更高的执行效率和性能，并具有良好的可行性。     

一种基于关系数据库的频繁项集挖掘算法

频繁项集的挖掘是数据挖掘中的一个十分重要的组成部分，目前对于事务数据库频繁项集的挖掘算法研究较多。本文根据事务数据库中布尔型频繁项集挖掘的理论和方法，再结合关系数据库的特殊性，利用标准SQL语言提出了一种新的在关系数据库中挖掘频繁项集的简易算法。实验证明该算法具有较高的效率。     

基于抽样的分布式约束性关联规则挖掘算法研究

本文采用抽样的方法，在基于约束的Eclat类算法（例如Eclat A和Eclat M）的基础上，提出了一种分布式约束性关联规则的挖掘算法——DMCASE算法。本算法在各数据站点上对一个较小的样本采用基于约束的Eclat类算法，挖掘局部约束频繁项集，采用归纳学习的方法归并所有局部约束频繁项集，产生全局约束频繁项集。只需1次扫描数据库，挖掘效率较高。实验证明：该算法是一种十分有效的解决基于约束条件下的分布式关联规则挖掘算法。     

频繁项集挖掘的研究与进展

挖掘频繁项集是许多数据挖掘任务中的关键问题，也是关联规则挖掘算法的核心，所以提高频繁项集的生成效率一直是近几年数据挖掘领域研究的热点之一，研究人员从不同的角度对算法进行改进以提高算法的效率。该文从频繁项集生成过程中解空间的类型、搜索方法和剪枝策略、数据库的表示方法、数据压缩技术等几个方面对频繁项集挖掘的基本策略进行了研究，对完全频繁项集挖掘、频繁闭项集挖掘和最大频繁项集挖掘的典型算法特别是最新算法进行了介绍和评述，并分析了各种算法的性能特点，指出其适于哪种类型的数据集。最后，对频繁项集挖掘算法的发展方向进行了初步的探讨。     

基于PFP-Growth算法的海量频繁项集挖掘

随着互联网技术的发展,网络数据变得越来越巨大,如何从中挖掘有效信息成为人们研究的重点。近年来频繁项集挖掘由于其在关联规则挖掘、相关挖掘等任务中的相关重要作用,越来越受到人们的重视。文中针对分布式计算环境下频繁项集挖掘算法的研究,对PFP-Growth算法进行了改进,通过MapReduce编程模型对改进的PFP-Growth算法进行了实现和应用,使用户可以从海量数据中高效地获得所有需要的频繁项集。实验结果表明算法在针对海量数据时具有较高的效率和伸缩性。     

基于动态项集计数的加权频繁项集算法

基于Apriori的加权频繁项集挖掘算法存在扫描数据集次数多的问题。为此,提出一种基于动态项集计数的加权频繁项集算法。该算法采用权值键树的数据结构和动态项集计数的方法,满足向下闭合特性,并且动态生成候选频繁项集,从而减少扫描数据集的次数。实验结果证明,该算法生成的加权频繁项集具有较高的效率和时间性能。     

一种高效的最大频繁项集挖掘算法DFMFI-Miner

分析最大频繁项集和完全频繁项集的关系，提出了一个挖掘最大频繁项集的高效算法DFMFI—Miner（The Miner Basedon Depth—First Searching for Mining Maximal Frequent Itemsets），采用深度优先方法搜索项集空间，采用垂直位图及一定的压缩方法对表示事务数据库并进行约简，并采用多种有效剪枝策略和优化策略，提高了算法的效率。在多个数据集上进行了实验，实验结果表明该算法特别适于挖掘具有长频繁项集的数据集。     

基于频繁项集挖掘算法的改进与研究

关联规则挖掘是数据挖掘领域中重要的研究内容,频繁项集挖掘又是关联规则挖掘中的关键问题之一。针对已有的频繁项集挖掘算法存在的问题,通过对Apriori算法的分析,提出了Inter-Apriori频繁项集挖掘算法。该算法使用交集策略减少扫描数据库的次数,从而使算法达到较高的效率。实验结果表明,Inter-Apriori算法是Apriori算法效率的2～4倍。     

利用抽样技术分布式开采可变精度的关联规则

关联规则是数据开采的重要研究内容,利用抽样及元学习技术提出一种快速的分布式开采可变精度的关联规则算法。为了能获得更准确的结果,还给出 采用适当缩小量小支持度和扩大全局检测的候选项集等技术的若干改进算法,最后给出了这种方法与类似方法的比较情况,算法具有效率高和通信量小的特点,尤适合效率比准确性要求更高的场合。     

Simulating the performance of SW-ARQ schemes within molecular communications

This paper provides results on an investigation concerning the application of five tailored Stop-and-Wait Automatic Repeat reQuest (SW-ARQ) schemes to a diffusion based molecular communication system. Each scheme is numerically simulated and evaluated to determine its performance with regards to average time cost and energy consumption. It is shown that all five schemes are beneficial depending upon the application scenario. Scheme 1 is the best choice for adjacent communications although, if a slightly higher energy budget can be afforded, schemes 2 and 3 will provide better performance than scheme 1 as the communication distance increases. Schemes 4 and 5 are designed to benefit scenarios with either a varying channel or for a channel with unknown parameters although will also benefit a static channel if again, further system energy can be utilised. This optimisation and trade-off between time and energy requirement for a complete successful transmission will become more important in future applications involving molecular communications where energy efficiency is a design consideration.     

并行量子遗传算法在QoS组播路由中的应用

随着网络通信技术的发展和Internet的普及,性能出色的组播路由越来越重要。著名的组播路由Steiner树问题是NP完全问题,应采用启发式方法求解。文中在常规量子遗传算法中引入并行进化模型,提出了一种解决多约束QoS组播路由优化问题的算法。在满足带宽、时延约束条件下寻找代价最小的组播树,并合理安排节点负荷,减少通信开销。仿真实验结果表明本算法搜索速度快、全局寻优能力强,性能和效率优于常规量子遗传算法。     

基于快速传输协议实现卫星TCP性能的改善

在经由卫星通信系统的网络应用中， TCP的性能较低。分段连接机制作为应用最广的一种性能优化方案，可以改善TCP端对端的性能，而在此机制中，卫星链路部分采用何种传输协议直接决定性能改善的效果。XTP协议是一种应用于有线网络的高速传输协议，但经过深入研究发现：XTP协议同样也非常适用于卫星通信系统。所以建议：在分段连接机制中，采用XTP协议作为卫星链路的传输协议。性能测试结果表明采用该建议非常显著地提高了端对端TCP的性能。     

An adaptive medium access control scheme for mobile ad hoc networks under self-similar traffic

An important function of wireless networks is to support mobile computing. Mobile Ad hoc NETworks (MANETs) consist of a collection of mobile stations communicating with each other without the use of any pre-existent infrastructure. The self-organization characteristic of MANETs makes them suitable for many real-world applications where network topology changes frequently. As a result, the development of efficient MAC (Medium Access Control) protocols in MANETs is extremely challenging. Self-similar traffic with scale-invariant burstiness can generate bursty network loads and thus seriously degrade the system performance. This paper presents an adaptive MAC scheme which dynamically adjusts the increasing function and resetting mechanism of contention window based on the status of network loads. The performance of this scheme is investigated in comparison with the legacy DCF (Distributed Coordination Function) under self-similar traffic and different mobility models. The performance results reveal that the proposed scheme is able to achieve the higher throughput and energy efficiency as well as lower end-to-end delay and packet drop probability than the legacy DCF.     

Efficient VSS free of computational assumption

VSS (verifiable secret sharing) is an important security mechanism in distributed computing. It is a secret sharing technique, where each distributed share holder can verify that he obtains a share of a unique secret. Two properties are important in VSS: efficiency and security, where the security of VSS includes two requirements, bindingness and hidingness. In any application of VSS, high efficiency is desired. In applications with critical security requirements, bindingness and hidingness are desired to be achieved without any computational assumption on hard mathematical problems. No existing VSS scheme is efficient enough in both computation and communication for cost-sensitive applications, especially in the popular ubiquitous and distributed computing environment. When security is required without any hard computational assumption, efficient VSS is even more difficult. To overcome this drawback in VSS, a new secret sharing protocol is proposed in this paper. It does not need any assumption or condition like hardness of computational problems to achieve bindingness and hidingness. Especially, its privacy is information-theoretic instead of statistical in most existing VSS schemes. It is very efficient in both computation and communication and costs much less than the existing VSS schemes. Therefore, it improves the efficiency of VSS and achieves the strongest security in VSS.     

一个有效的Join任务图的调度算法

已有的Join任务图的调度算法大多不是基于通信竞争的环境而开发,且未考虑节省处理机的问题,使算法的应用效果不佳.因此,针对Join任务图,提出一个通信竞争环境的调度算法,该算法因串行通信边而改善其调度效率,时间复杂度为O(vlogv),其中,v为图中任务的个数.实验结果表明,与其他算法相比,该算法的调度长度较短且使用的...     

Predicting communication protocol performance on superscalar architectures using instruction dependency

Increasing diversity in telecommunication workloads leads to greater complexity in communication protocols. This occurs as channel bandwidth rapidly increases. These factors result in larger computational loads for network processors that are increasingly turning to high performance microprocessor designs. This paper presents an analytical method for estimating the performance of instruction level parallel (ILP) processors executing network protocol processing applications. Instruction dependency information extracted while executing an application is used to calculate upper and lower bounds for throughput, measured in instructions per cycle (IPC). Results using UDP/TCP/IP applications show that the simulated IPC values fall between the analytically derived upper and lower bounds, validating the model. The analytical method is much less expensive than cycle-accurate simulation, but reveals similar throughput performance predictions. This allows the architectural design space for network superscalar processors to be explored more rapidly and comprehensively, to reveal the maximum IPC that is possible for a given application workload and the available hardware resources.     

Adapting wave-front algorithms to efficiently utilize systems with deep communication hierarchies

Large-scale systems increasingly exhibit a differential between intra-chip and inter-chip communication performance especially in hybrid systems using accelerators. Processor-cores on the same socket are able to communicate at lower latencies, and with higher bandwidths, than cores on different sockets either within the same node or between nodes. A key challenge is to efficiently use this communication hierarchy and hence optimize performance. We consider here the class of applications that contains wave-front processing. In these applications data can only be processed after their upstream neighbors have been processed. Similar dependencies result between processors in which communication is required to pass boundary data downstream and whose cost is typically impacted by the slowest communication channel in use. In this work we develop a novel hierarchical wave-front approach that reduces the use of slower communications in the hierarchy but at the cost of additional steps in the parallel computation and higher use of on-chip communications. This tradeoff is explored using a performance model. An implementation using the reverse-acceleration programming model on the petascale Roadrunner system demonstrates a 27% performance improvement at full system-scale on a kernel application. The approach is generally applicable to large-scale multi-core and accelerated systems where a differential in communication performance exists.     

Juggling on a high wire: Multitasking effects on performance

In this study, we develop a theoretical model that predicts an inverted-U relationship between multitasking and performance. The model is tested with a controlled experiment using a custom-developed application. Participants were randomly assigned to either a control condition, where they had to perform tasks in sequence, or an experimental condition, where they could discretionarily switch tasks by clicking on tabs. Our results show an inverted-U pattern for performance efficiency (productivity) and a decreasing line for performance effectiveness (accuracy). The results of this study indicate that the nature of the relation between multitasking and performance depends upon the metric used. If performance is measured with productivity, different multitasking levels are associated with an inverted-U curve where medium multitaskers perform significantly better than both high and low multitaskers. However, if performance is measured with accuracy of results, the relation is a downward slopping line, in which increased levels of multitasking lead to a significant loss in accuracy. Metaphorically speaking, juggling multiple tasks is much more difficult while balancing on a high wire, where performance mishaps can have serious consequences.