共查询到19条相似文献,搜索用时 78 毫秒
1.
2.
用户位置感知的Web服务QoS预测方法 总被引:1,自引:0,他引:1
随着互联网上Web服务数量的增多,服务的QoS开始成为用户在选择服务时关注的焦点.为了有效获取服务的QoS信息以便向用户推荐高质量的服务,提出一种用户位置感知的Web服务QoS预测方法--UL-WSRec.利用一些QoS属性受用户位置影响的事实和同一个自治系统中的用户在位置上邻近的特点,根据自治系统来计算用户位置和划分相似用户.在传统的协同过滤算法基础上,利用用户位置信息提出了改进的算法,以提高服务QoS预测的效率和精度.基于真实Web服务数据的实验表明了该方法的有效性. 相似文献
3.
数据库技术在模糊聚类预测中的应用研究 总被引:7,自引:1,他引:6
将数据库技术应用于模糊聚类预测中,用FOXPRO开发了模糊聚类预测软件,通过对上海水泥厂提供的实测数据的实旆计算,取得了令人满意的结果,文中提供的方法具有普遍的意义和广阔的应用前景,作为数据库技术原应用开辟了新领域。 相似文献
4.
如何为用户准确预测Web服务的QoS是最近的一个热点研究问题.考虑到Web服务的QoS可能受用户具体位置的影响,一些工作提出了位置感知的Web服务QoS预测方法.然而,很少有工作在实践中针对Web服务QoS与用户位置的相关性进行具体验证.提出了一套对Web服务QoS与用户位置的关系进行验证的方法,并使用真实的Web服务QoS数据集进行了实验.实验与分析发现:用户位置越邻近,他们在相同Web服务上观察到的QoS越相似;反过来也是如此.这些结果表明Web服务QoS与用户位置之间具有较强的相关性,可以为位置感知的Web服务QoS预测方法提供很好的依据. 相似文献
5.
移动对象位置预测是基于位置服务的重要组成部分。现有的移动对象位置预测算法有基于马尔可夫链的算法、基于隐马尔可夫模型的算法、基于神经网络的算法等,然而这些算法都无法解决移动对象轨迹数据中位置过多带来的维数灾难问题。为了解决这一问题,提出了位置分布式表示模型(location distributed representation model,LDRM)。该模型将难以处理的表示位置的高维one-hot向量降维成包含移动对象运动模式的低维位置嵌入向量。随后,将该模型与基于长短期记忆网络(long short-term memory,LSTM)的位置预测算法结合为LDRM-LSTM移动对象位置预测算法。真实数据集上的实验表明,与现有算法相比LDRM-LSTM算法在预测准确性上有较大的提升。 相似文献
6.
基于多项式拟合法的空中目标实时位置预测研究* 总被引:1,自引:0,他引:1
李亚宁 《计算机与数字工程》2015,(3):404-407,411
针对基于G PS的空中目标跟踪系统中机载G PS位置信息易受干扰,目标位置信息不能连续实时获取的问题,提出采用基于最小二乘的多项式拟合法进行目标位置的实时预测,根据预测值对目标进行跟踪定位。并对该方法在实际应用过程中的参数选择问题进行了仿真分析,给出了应用参考。通过实际数据分析表明,用该方法进行目标位置预测,具有实时性强,精度高的优点,并可在数据丢失的情况下,对目标位置信息进行有效补偿。 相似文献
7.
准确预测用户的地理位置可以有效地改善基于位置服务的质量.针对标准Markov模型预测能力不足,以及多阶Markov模型阶数难以确定的问题,提出了一种基于自适应多阶Markov模型的位置预测方法.该方法采用一种基于规则图形的方式对原始位置信息进行抽象化处理,并使用一种基于训练数据的启发式方式自动确定用于预测的模型阶数.最后,基于真实的位置数据,对自适应多阶Markov模型的预测性能进行了评测.结果表明自适应多阶Markov模型的预测精度和预测长度始终高于多阶Markov模型,平均预测精度提高将近20%,平均预测长度提高将近10个单位区域,且不易受训练数据质量影响. 相似文献
8.
9.
在网页制作中,我们常会遇到这样的问题:图片分辨率一旦变大、窗口最大化之后,图层就错位了。而使用菜单插入层的时候,这个层就会与它的上级元素相对定位,一旦移动它,它就会马上变成绝对定位。这个问题虽小,但在实际应用过程中还是会给网页设计者带来一些麻烦。 相似文献
10.
11.
12.
为了改善基于位置的服务中查询迟缓和信息失效等问题,提出了数据预处理技术.结合移动互联网特点,通过服务器端数据预处理和提前发送,使部分移动终端向服务器端数据库的查询转化为移动终端上本地数据库查询,从而减少等待时间;运用智能学习策略提高服务信息命中率,使其更贴近用户的实际需求.实验结果表明,预处理技术能有效地提高查询效率,降低对不稳定无线网络的依赖性,减少数据的出错和重发. 相似文献
13.
14.
Jae-Kwan Yun Dong-Oh KimDong-Suk Hong Moon Hae KimKi-Joon Han 《Computers & Industrial Engineering》2006
Recently, as the growth of the wireless Internet, PDA, and HPC, the focus of research and development has been changed to LBS (location based service). To offer LBS efficiently, there must be a real-time GIS platform that can deal with dynamic status of moving objects and a location index which can deal with the characteristics of location data. Therefore, in this paper, we studied the real-time mobile GIS based on the HBR-tree to manage mass of location data efficiently. The real-time mobile GIS mainly consists of the HBR (Hash Based R)-tree and the real-time GIS platform. The HBR-tree is a combined index type of the R-tree and the spatial hash index. Although location data are updated frequently, update operations are done within the same hash table in the HBR-tree, so it costs less than other tree-based indexes. The real-time GIS platform consists of a real-time GIS engine, a middleware, and a mobile client. 相似文献
15.
16.
17.
当前的移动推荐系统只将位置信息作为推荐属性处理,弱化了其在推荐中所起的作用,更重要的是忽略了移动生活服务位置相关性和用户空间运动有界性特征。针对该问题,设计了基于位置簇的用户偏好表示模型和移动生活服务个性化推荐算法。该算法通过模糊聚类得到位置簇,使用遗忘因子调节用户在该位置簇对服务资源属性值的偏好,并且采用概率分布和信息熵理论计算属性权重,按位置簇对用户偏好和服务资源进行匹配得到top-N推荐集。由于位置簇的定义,使得算法给出与用户偏好相似度较高的服务资源。案例分析结果符合这一结论,从而验证了算法的有效性和精确性。 相似文献
18.
An important privacy issue in Location Based Services is to hide a user’s identity while still provide quality location based
services. Previous work has addressed the problem of locational -anonymity either based on centralized or decentralized schemes. However, a centralized scheme relies on an anonymizing server
(AS) for location cloaking, which may become the performance bottleneck when there are large number of clients. More importantly,
holding information in a centralized place is more vulnerable to malicious attacks. A decentralized scheme depends on peer
communication to cloak locations and is more scalable. However, it may pose too much computation and communication overhead
to the clients. The service fulfillment rate may also be unsatisfied especially when there are not enough peers nearby. This
paper proposes a new hybrid framework called HiSC that balances the load between the AS and mobile clients. HiSC partitions
the space into base cells and a mobile client claims a surrounding area consisting of base cells. The number of mobile clients
in the surrounding cells is kept and updated at both client and AS sides. A mobile client can either request cloaking service
from the centralized AS or use a peer-to-peer approach for spatial cloaking based on personalized privacy, response time,
and service quality requirements. HiSC can elegantly distribute the work load between the AS and the mobile clients by tuning
one system parameter base cell size and two client parameters - surrounding cell size and tolerance count. By integrating salient features of two schemes, HiSC successfully preserves query anonymity and provides more scalable and
consistent service. Both the AS and the clients can enjoy much less work load. Additionally, we propose a simple yet effective
random range shifting algorithm to prevent possible privacy leakage that would exist in the original P2P approach. Our experiments
show that HiSC can elegantly balance the work load based on privacy requirements and client distribution. HiSC provides close
to optimal service quality. Meanwhile, it reduces the response time by more than an order of magnitude from both the P2P scheme
and the centralized scheme when anonymity level(value of ) or number of clients is large. It also reduces the update message cost of the AS by nearly 6 times and the peer searching
message cost of the clients by more than an order of magnitude.
Chengyang Zhang received his B.S. degree in Industrial Automation from University of Science and Technology, Beijing in 2000 and master degree in computer system engineering from University of Science and Technology, Beijing in 2003. He is currently a Ph.D. student at the Computer Science and Engineering department of University of North Texas, Denton, TX, USA. His research interests include location based services, spatio-temporal databases, and geo-stream data management systems. Yan Huang received her B.S. degree in Computer Science from Beijing University, Beijing, China, in July 1997 and Ph.D. degree in Computer Science from University of Minnesota, Twin-cities, MN, USA, in July 2003. She is currently an assistant professor at the Computer Science and Engineering Department of University of North Texas, Denton, TX, USA. Her research interests include geo-sensor networks, spatial databases, and data mining. She is a member of the IEEE Computer Society, the ACM, and the ACM SIGMOD. Her research is supported by Texas Advanced Research Program (ARP), Oak Ridge National Lab, and NSF. 相似文献
Chengyang ZhangEmail: |
Chengyang Zhang received his B.S. degree in Industrial Automation from University of Science and Technology, Beijing in 2000 and master degree in computer system engineering from University of Science and Technology, Beijing in 2003. He is currently a Ph.D. student at the Computer Science and Engineering department of University of North Texas, Denton, TX, USA. His research interests include location based services, spatio-temporal databases, and geo-stream data management systems. Yan Huang received her B.S. degree in Computer Science from Beijing University, Beijing, China, in July 1997 and Ph.D. degree in Computer Science from University of Minnesota, Twin-cities, MN, USA, in July 2003. She is currently an assistant professor at the Computer Science and Engineering Department of University of North Texas, Denton, TX, USA. Her research interests include geo-sensor networks, spatial databases, and data mining. She is a member of the IEEE Computer Society, the ACM, and the ACM SIGMOD. Her research is supported by Texas Advanced Research Program (ARP), Oak Ridge National Lab, and NSF. 相似文献
19.
针对在线视频热度预测研究中分类及预测效果欠佳,规则化较多和较缺乏实践检验等问题,通过对实际在线视频服务系统所采集的海量数据研究,提出一种基于深度信念网络(Deep Belief Networks,DBNs)的视频热度预测方法。首先,结合社交网络的关注度和视频关键词的搜索热度,对影响因子进行了建模和量化处理;其次,根据输入和输出变量确定了DBNs各层网络的结构,优化了网络参数和预测模型;最后,通过在线视频服务商的数据对深度信念网络进行训练,并多次交叉实验对比分析,结果表明基于DBNs方法在视频热度预测上准确率最高79.47%(国内视频)、65.33%(国外视频),可以为在线视频上映前的投资、宣传以及风险评估提供较全面可靠的参考决策。 相似文献