共查询到20条相似文献,搜索用时 468 毫秒
1.
提出一种将局部近邻传播和考虑用户特征的相似性测度相结合实现社交网络中的社区识别的算法。一方面,通过放松代表点约束条件及限制消息传播范围为节点的局部近邻,算法在降低时间和空间复杂度的同时保持较小的识别精度损失,从而能够适应社交网络挖掘需要;另一方面,通过将节点的拓扑相似度和特征相似度相结合来描述节点的综合相似度,使算法能够适应社交网络采样数据中用户关联信息不完整的情况。通过在人工数据集和真实数据集上的对比实验表明,所提方法不仅具有近似线性的时间复杂度及线性的空间复杂度,而且在网络中的节点关联边信息不完整时仍保持较好的识别精度。 相似文献
2.
3.
针对目前电子取证缺乏全程监督的问题提出了一种基于用户行为关联分析的电子取证系统.该系统根据业务活动主体、行为和客体约束关系建立用户行为知识库,基于电子证据属性相似度,对原始电子证据进行过滤和融合,在WINEPI算法的基础上,实现电子证据的用户行为关联.使用实际数据测试表明,该系统能够对用户网络活动过程进行全程监督,并可以把取证过程的监督数据作为呈堂的证据. 相似文献
4.
采用交互式遗传算法求解大数据信息检索问题时,为实现偏好信息的提取和优化,单用户需完成较多数量的人-机交互操作,由此易产生用户疲劳、算法搜索效率低的难题。对此,该文在算法中引入多用户并行策略,通过群体决策优势,提高样本利用效率。首先,根据优化目标性质确定共性化协同或个性化协同类型,基于用户浏览行为计算用户相似度和个体相似度。然后,通过共享偏好相似用户的偏好相似个体预测个体区间适应值。基于个体表现型相似度聚类,提出大规模种群个体“区间数-区间数”适应值赋值策略。最后,依据子代种群个体与父代种群最优个体的相似性,推荐用户最佳评价个体。将所提方法应用于装饰性墙壁纸选型问题,并与已有典型方法比较。结果表明,所提方法在推荐个体质量、减轻用户疲劳、提高搜索效率等方面均具有优越性。 相似文献
5.
基于特征事务元素的用户事务聚类方法研究 总被引:4,自引:2,他引:2
针对现有用户访问事务聚类算法复杂度较高的问题,文章提出了一种新的用户事务聚类算法。首先对用户事务中访问过的页面浏览次数和时间进行综合统计,并定义了函数,将其映射为页面访问时间属性值;其次通过抽取事务的特征事务元素集,建立新的聚类对象集事务,实现对原有事务的降维。最后给出计算事务相似度的方法.并得到一个模糊相似度矩阵以实现事务聚类。该算法具有较高的准确性,计算复杂度低,扩展性好,可以广泛地应用于电子商务领域。 相似文献
6.
7.
该文采用权重增量及相似聚集的用户行为分析算法,为用户推荐个性化视频提供了一个有效的解决方案。方法包含3个主要部分,首先利用RFM(Recentness, Frequency, Monetary amount)模型分析用户的行为,将相同行为的用户归为一组;然后结合用户的最近习惯,使用基于权重增量的Apriori算法挖掘用户之间的关联规则,并用向量空间模型进行相似度计算从而实现用户相似聚集;最后进行协同过滤式推荐,完成整体个性化视频推荐过程。该方法的特点是行为数据自动收集获取,避免了直接对视频大数据的处理;另外,视频推荐随着用户行为的改变而动态变化,更加符合实际情况。实验结果表明,该方法有效并且稳定,相比于单一推荐方法,在准确率、召回率等综合指标上均有明显提升。 相似文献
8.
9.
10.
11.
12.
The degradation of end-to-end quality of service (QoS) on mobile users is becoming a common issue for IEEE 802.11 infrastructure-based networks in crowded areas. This research tackles the issue by employing an SDN framework on an integrated wireless/wired environment. Thereby, we present the development and implementation of a system that performs user management by analyzing the network load from the OpenFlow statistics, as well as the wireless information collected from the associated users. In order to analyse the behaviour of the proposed user migration algorithm, we evaluate the system under scenarios with different traffic load and user session duration. From the experiments, we observed that in several cases wireless users get a considerable QoS improvement at the application layer (up to 30% improvement in throughput and up to 20% in delay in our simulations) once the system is activated. Based on the results, we present an analysis on how and when user migration in multi-access point IEEE 802.11 networks can be most effective. 相似文献
13.
结合关联规则与模糊C均值聚类(Fuzzy C-Means,FCM)进行用户聚类,首先将用户访问事务集构造出页面关联矩阵,以此来计算用户浏览路径之间的相似程度,接着利用FCM算法对用户进行聚类.通过与传统聚类比较的实验证明,此方法是有效的且更符合实际. 相似文献
14.
用户分组算法作为非正交多址接入(NOMA)的关键部分,对系统吞吐量和用户公平性具有重要影响。当用户数量和可用的资源增加时,用户分组的最佳调度将变得不可行,该文提出一种子带间多用户分组优化算法。该算法首先根据用户信道增益差异和子带复用用户数限制条件进行多用户初始分组处理,减小用户搜索空间,然后逐步完成初始分组用户间的优化组合,以几何平均用户吞吐量最大为用户分组准则,进一步提升小区边缘用户的吞吐量。仿真结果表明,所提算法与传统的用户分组算法相比,系统总吞吐量和几何平均用户吞吐量性能提升均超过3%。 相似文献
15.
16.
Blind Spectrum Sensing Algorithms for Cognitive Radio Networks 总被引:2,自引:0,他引:2
《Vehicular Technology, IEEE Transactions on》2008,57(5):2834-2842
17.
Siavash Ghavami Bahman Abolhassani 《International Journal of Communication Systems》2012,25(2):121-145
In this paper, a two‐phase algorithm for the spectrum sensing and power/rate control of a secondary user (S‐user) or cognitive radio is proposed. In the first phase, the primary base station (P‐BS), which is conscious of both the number and the data rate of primary active users (P‐user), broadcasts theitusage capacity percentage (UCP) of its cell. Since knowing only the UCP is not enough to guarantee that the total load (of P‐users and S‐users) is less than a maximum permissible load, the S‐user must measure the total interference received from both the P‐users and other S‐users. In this direction, using both the UCP and measurement of the interference received from the P‐users and the S‐users by the S‐user or secondary base station (S‐BS), we mathematically derive an equation for issuing data transmission permission, which if it is held then the second phase of the algorithm: the transmit power/rate control starts. In this phase, the S‐user and the S‐BS look for feasible values for the transmit power level and transmission rate. If there are feasible values, it starts its transmission at these feasible transmit power and rate. Since both the location of the S‐user and the channel condition vary in time, the whole algorithm is iterated periodically with a period faster than the coherence time of the channel. Furthermore, we consider the down link of the above system with cooperation among neighboring S‐users to overcome fading channels and we investigate the amount of improvement in the reliability of the issuing data transmission permission. As well, we consider the uplink of the system with multiple antennas in the S‐BS to investigate the improvement in the same parameter over spatially correlated and independent fading channels. Theoretical analysis is validated using computer simulations. Both theoretical analysis and computer simulations show that the proposed cooperative spectrum sensing algorithm performs properly at SNR = ?5dB in flat Nakagami‐m fading channels with m = 1 even in correlated fading channels. We also address the improvement of the reliability of the issuing data transmission permission in the uplink in case of using multiple antennas only in the S‐BS. Copyright © 2011 John Wiley & Sons, Ltd. 相似文献
18.
为改善用户自身天线间的干扰和信道噪声干扰对SLNR算法性能的影响,对SLNR算法进行了改进,包括两种改进算法:SLNR_ZF算法是利用ZF算法可以消除用户自身天线间干扰的作用将其引入到SLNR算法中;SLNR_MMSE算法是在SLNR_ZF算法的基础上引入规范化因子,消除噪声干扰。通过仿真结果表明,改进的两种优化算法可以有效地消除包括用户自身天线间干扰、用户间干扰和噪声干扰的多种干扰,提高小区边缘用户的吞吐量和误码率。 相似文献
19.
随着无线通信业务的不断增长,频谱资源越来越紧缺,然而另一方面大量授权的无线频谱却被闲置或者利用率极低,于是认知无线电技术应运而生,已成为无线通信领域的研究热点。认知无线电的基本思想是次用户(认知用户)利用主用户(授权用户)未占用的空闲频谱进行通信,其可用无线资源是根据授权用户的频谱使用情况而动态变化的。因此,能否实现对系统可用无线资源的合理有效管理,对整个认知无线电系统性能的优劣起着决定性作用。本文提出了一种在干扰温度限制下基于公平的功率与信道联合分配算法,该算法在主用户干扰温度及次用户发射功率的双重限制下,以最大化系统容量为基本目标,实现信道与功率的联合分配,并且引入贫困线来保证各个用户信道分配的公平性。论文建立了该问题的非线性规划数学模型,给出了模型的求解方法,并进一步设计了具体分配算法及其步骤。论文对干扰门限分别为-90dBm、-95dBm、-100dBm、-105dBm、-110dBm时的系统归一化容量累积分布函数进行了仿真比较,发现当干扰门限越低时,本文算法的优势越明显。这是因为在干扰门限较低时,干扰温度限制是功率分配的主要制约因素,而本文的算法正是基于干扰门限进行分配的。因此基于干扰温度限制的公平的功率与信道联合分配算法具有良好的性能,在保证了系统的公平性效益的同时,提高了系统的归一化容量。 相似文献
20.
Application deployment in sessions composed of several users is now a hot topic. These users communicate together with heterogeneous terminals. Deployed applications on these nodes must fit to the execution environment and must be interoperable with applications already installed on the others nodes of the session. In this paper, we propose an architecture, which provides a user with missing applications according to the session requirements while respecting compatibility and interoperability constraints. This decentralized and distributed architecture is based on a context-aware deployment algorithm running on each node. After discovering applications scattered on a peer-to-peer network, the algorithm generates deployment configurations needed for any deployment node. Then, the algorithm performs the necessary downloads and instantiations. We present our context-aware deployment platform composed of generic modules. These modules includeApis to build deployment services according to this architecture. 相似文献