一种基于网页浏览行为的用户兴趣度计算方法

用户兴趣度的计算是个性化服务的核心,而用户浏览网页时的行为能反映用户的兴趣。文中着重分析保存页面、收藏页面以及网页浏览速度这三种网页浏览行为,且在此基础上结合用户兴趣衰减因子,提出一种新的基于网页浏览行为的用户兴趣的计算方法,并利用K-means算法将浏览的网页内容聚类成不同主题,最后基于这三种网页浏览行为和用户兴趣衰减因子计算用户主题兴趣度。实验结果表明,提出的用户兴趣度计算方法是有效的。     

基于隐语义模型改进的协同过滤推荐算法

为解决传统的协同过滤算法相似度计算不精确而导致的推荐召回率低的问题。通过LFM隐语义模型与协同过滤结合的方法研究了用户不相似度与用户平均兴趣程度对推荐召回率的影响,并且提出依据个人选择能力的个性化推荐策略。结果表明:该算法能够有效优化用户间的相似度,提高推荐结果的召回率。     

面向MANET环境的动态自适应副本放置算法研究

移动自组网环境给数据复制技术带来了新的挑战。本文提出了一种分布式动态自适应副本放置算法,算法以数据访问的通信开销作为目标函数,每个副本节点从其邻居节点收集读写请求的局部信息,动态地调整副本放置方案以适应访问请求和网络拓扑的变化。本文还提出了基于稳定邻居的副本放置算法以减小由于节点移动给副本放置带来的振荡。模拟实验表明,在移动自组网环境中,本文提出的算法有效地减少了数据访问的通信开销,提高了数据访问的效率。     

基于多兴趣度的图书借阅推荐系统研究与设计

针对数字图书馆图书借阅推荐系统中通常采用的协同过滤算法中存在的不足,分析了该算法存在在用户多兴趣情况下推荐时难免会产生偏差,从而降低预测推荐准确率问题,进而提出了一种基于用户多兴趣度的图书借阅推荐系统推荐算法的设计思路。该算法首先对待推荐图书进行分类,然后根据用户的借阅情况计算其对每类图书的兴趣度,将对每类图书兴趣度相似的用户聚为一类,在小范围内寻找最近邻居,最后根据用户对不同类别书目的兴趣度作为权重值来产生对该用户的每类图书的推荐数目。该算法不仅能够解决用户多兴趣度问题,同时也可减少计算工作量,提高推荐的准确率。     

基于主题的用户兴趣域算法

针对用户兴趣偏好多变问题,提出一种兴趣特征权重随时间而变化的迭代计算方法。构造了用户兴趣特征与主题类间的二部图关系,并在此基础上提出了一种基于主题的用户兴趣聚类算法(TBC),改变了聚类对象"非此即彼"的硬划分方式。该算法所形成的基于主题的用户兴趣域结构,不仅充分表达了用户的多域兴趣特征和域间主题的联系,还能适应用户兴趣变化。实验表明,TBC算法比传统的K-Means算法以及属于软划分方式的FCM聚类具有更好的用户兴趣划分效果,并且在个性化推荐服务中表现出更高的推荐质量和效率。     

基于本体与模式的网络用户兴趣挖掘

本文探讨了用户兴趣挖掘的新方法,首先从用户搜索日志中获取访问行为元素,并借助通用本体中的概念描述网页所体现的用户个体兴趣,然后提出了一种兴趣得分计算方法,并在此基础上从用户个体兴趣序列中识别不同的兴趣模式,判断用户的短期兴趣,并利用通用本体得出用户兴趣的集合表示,最后根据短期兴趣的增量积累推算长期兴趣.整个过程避开了以往兴趣挖掘方法中通过相似度计算和文档聚类算法进行兴趣合并的问题,为兴趣发现提供了新思路.实验结果表明,本文的方法对用户兴趣的描述更具体,取得了更优化的兴趣合并结果.     

基于累积点播概率的负载均衡算法研究

为了提高集群系统的资源利用率,降低对用户请求的拒绝率,需要对集群系统中的各个服务器节点进行负载均衡.文章在对一些主流的负载均衡算法进行了详细分析,对比了它们的优缺点,并在此基础上提出了基于累积点播概率的负载均衡算法.该算法包含三个部分:最优副本生成、贪婪的副本放置和动态贪婪副本调整.仿真实验验证了所提算法的有效性.     

一种基于用户浏览行为更新的兴趣模型

为了使个性化推荐系统能适应用户浏览兴趣的变化,一个关键问题就是如何动态地更新用户的兴趣模型。提出了一种新的二层树状结构模型。模型的第1级节点表示用户的兴趣类型,及其对每一兴趣类的兴趣度高低：第2级节点是用户某一兴趣类的向量空间模型表示。并通过分析Web环境下用户的浏览速度,改进了兴趣类兴趣权值计算公式,然后通过用户在这张页面上的浏览速度和页面内容与兴趣类主题的相关程度来更新模型。试验表明改进的计算公式能将兴趣估计的一致性提高到70％。     

一种基于用户浏览行为更新的兴趣模型

为了使个性化推荐系统能适应用户浏览兴趣的变化,一个关键问题就是如何动态地更新用户的兴趣模型.提出了一种新的二层树状结构模型.模型的第1级节点表示用户的兴趣类型,及其对每一兴趣类的兴趣度高低;第2级节点是用户某一兴趣类的向量空间模型表示.并通过分析Web环境下用户的浏览速度,改进了兴趣类兴趣权值计算公式,然后通过用户在这张页面上的浏览速度和页面内容与兴趣类主题的相关程度来更新模型.试验表明改进的计算公式能将兴趣估计的一致性提高到70%.     

基于LDA用户兴趣模型的远程教育课程推荐方法研究

传统的远程教育课程推荐方法因数据稀疏问题,造成其主题集中性较差,为此设计基于LDA用户兴趣模型的远程教育课程推荐方法。通过远程教育课程外在属性包容度和内在属性质量值,计算远程教育课程的重要度,并以重要度为依据,利用LDA用户兴趣模型判断用户对主题的偏好度,确定主题与远程教育课程的相似度系数,获得用户对远程教育课程的兴趣度,以此为基础完成远程教育课程的推荐。实验结果表明:使用基于LDA模型的推荐方法向用户推荐的课程有50%以上都是用户需求的课程,而传统的推荐方法只有不到20%,两者相比,基于LDA模型的推荐方法的主题集中性更强,更适合应用在远程教育课程推荐中。     

基于预期延迟值的扩散转发路由算法

针对现有延迟容忍网络路由算法在消息传递过程中分发不灵活的问题,提出基于节点通话历史记录计算预期通断时长以及消息传递延迟值,并根据结果选择节点、分配副本数进行消息转发的路由算法（MDESF）。在算法不同阶段,根据节点的活跃度选择中继节点,再分配传递消息的副本数目,加快分发,降低延迟。同时为保证消息及时传递,在单副本阶段选择最优节点转发。本算法同时提出内存管理策略,提高消息转发成功率。仿真实验表明,该算法相对其他同类算法在投递率、交付时延和网络拥塞比率方面综合表现更优。     

基于内容的音乐检索算法研究

基于内容的音乐检索把用户的哼唱作为输入,搜索更加智能,是目前音乐检索研究的热点和重点。在深入研究已有音乐检索算法的基础上．提出了基于基频矩阵的基音检测算法,有效提高了背景音乐下基音提取的准确性。对DTW算法进行了改进,引入高度动态调整和可变长搜索技术,提高了哼唱输入与音乐库中音乐匹配的性能。     

Efficient algorithms for scheduling data broadcast

With the increasing acceptance of wireless technology, mechanisms to efficiently transmit information to wireless clients are of interest. The environment under consideration is asymmetric in that the information server has much more bandwidth available, as compared to the clients. It has been proposed that in such systems the server should broadcast the information periodically. A broadcast schedule determines what is broadcast by the server and when. This paper makes the simple, yet useful, observation that the problem of broadcast scheduling is related to the problem of fair queueing. Based on this observation, we present a log‐time algorithm for scheduling broadcast, derived from an existing fair queueing algorithm. This algorithm significantly improves the time‐complexity over previously proposed broadcast scheduling algorithms. Modification of this algorithm for transmissions that are subject to errors is considered. Also, for environments where different users may be listening to different number of broadcast channels, we present an algorithm to coordinate broadcasts over different channels. Simulation results are presented for proposed algorithms.     

Downlink scheduling with guarantees on the probability of short-term throughput

We consider the problem of scheduling multiple transmissions on the downlink of a wireless network with performance guarantees in the form of the probabilities that short term throughputs exceed user specified thresholds. Many interactive data applications have some degree of a latency requirement, and measure performance by throughput over a relatively short time interval. We refer to the fraction of time such user throughput reaches a predefined rate threshold or higher as tail probability. The problem is formulated as maximizing the minimum ratio of tail probability to the user specified probability threshold. We present necessary and sufficient optimality conditions for the case in which the time interval of interest is consistent with the time scale of channel variation. An online algorithm is proposed which can achieve the optimality. For the case in which the time interval of interest is large compared to the time scale of channel variation, we develop an online algorithm which attempts to maximize the minimum normalized tail probability by taking the advantage of channel variation over users and over time. Simulation results demonstrate that the proposed algorithm can achieve better performance than other algorithms such as the proportional fair algorithm and the Max C/I algorithm.     

Time-variant interest community based query message routing algorithm in opportunity social network

To solve the problem that the successful rate of resource query is low in opportunity social network,a query message routing algorithm based on time-variant interest community was proposed.Two well-known mobile social datasets were firstly analyzed,and then the temporal and spatial correlations of users were found.The time-variant interest communities based on temporal and spatial correlations of users were constructed.Next,a dynamic query routing on the constructed time-variant interest communities was proposed.Theoretical analysis shows that the proposed routing can run in O (nlog n) time.The comparisons between the proposed routing and state-of-the-art message delivery algorithms show that the proposed routing can keep high query success rate,reduce the average query latency and the hop count of a query and maintain low system overhead.     

一种融合用户偏好和社交活跃度的推荐算法

为有效解决传统推荐算法精度低的问题,提出了一种融合用户偏好和社交活跃度的概率矩阵分解推荐算法(Probabilistic Matrix Factorization Recommendation Algorithm Combining User Prefer-ence and Social Activity,UPSA-P...     

TSRN: A Tabu Search-Based Routing Protocol for Named Data Networking

Information-centric network (ICN) has become a basic research topic with regard to the future architecture of the Internet. Storage and routing data is an important feature of ICN, as it improves network performance and content distribution. One of the challenges in this type of network is the routing problem. The objective of this paper is to propose a routing algorithm in the named data network. Owing to the impressive results presented by the enhanced heuristic methods, tabu search has been selected to develop the proposed algorithm. This work has been carried out in five major steps, Creating the initial solution, The mechanisms of generating a neighborhood, Evaluating the neighbor solutions and selecting the appropriate responses, Updating the answer and the tabu list, Controlling the algorithm termination conditions. The proposed protocol has been implemented in the simulator NS3. Important measures such as the packet delivery ratio, satisfied the interest packets ratio, the average number of interest packets, end-to-end delay, route creation delay, transmission delay, loss rate, and control message overhead have been studied. These show the quality of the proposed protocol. To evaluate the performance of the proposed algorithm, the results of the algorithm were compared with OSPFN and OSPF, which are the most well-known algorithms. The results of this assessment of throughput, control message overhead, packet delivery ratio, and end-to-end delay show that the proposed algorithm has the ability to provide appropriate solutions comparable to these algorithms.     

Network real-time scheduling algorithm based on multi-feature dynamic priority

Real-time task scheduling system structure and task model were proposed aiming at the network real-time scheduling problem.The task degree of urgency was defined by considering the deadline of task,execution time and interval time between works.The task degree of tightness was proposed based on service-level assurance,according to functional importance of different tasks in the real-time task scheduling system.The thrashing limit for avoiding task switching frequently was acquired through dynamic regulation to task priorities by degree of urgency and degree of tightness,which guaranteed the success rate of tasks execution and utilization ratio of client execution.Test simulation results suggest that the multi-feature dynamic priority scheduling strategy improves the success rate of task scheduling and shorten the average response time,which suggests it has obvious superiority compared with BE and EDF scheduling algorithm.     

一种目标响应自适应的通道可靠性跟踪算法

为解决基于时空正则项的目标跟踪算法(STRCF)在目标短时遮挡时定位精度低和目标旋转时尺度估计不准确的问题，该文提出了一种目标响应自适应的通道可靠性跟踪算法。该算法在目标模型训练时加入了目标响应正则项，通过在求解过程中更新理想目标响应函数，使得目标被短时遮挡后可重新跟踪目标；加入通道可靠性评价各特征通道的可靠性，提高了模型对目标的表达能力；将目标图像转换至对数极坐标系下训练尺度滤波器，提高在目标旋转时的尺度估计精度。实验结果表明，该文所提算法较STRCF在平均中心位置误差中降低了28.54个像素，在平均重叠率中提高了22.8%，在OTB2015数据集下成功率曲线下面积较STRCF提高了1.5%。     

基于自私行为分析的超密集D2D中继选择算法

在超密集D2D通信中,目前的中继选择方案主要假定中继设备具有主动意愿参与数据转发,但部分理性中继设备出于自身考虑,可能会存在时延或拒绝转发的自私行为,进而影响用户体验效果.从中继设备自私行为角度出发,提出了D2D通信中自私中继设备识别方法,进而提出了基于自私行为分析的超密集D2D中继选择算法.首先,定义了联合兴趣度、转...