检索结果-内蒙古大学图书馆

Incremental mining of sequential patterns from multiple item set data streams

ICIC Express Letters 2014年第10期8卷 2687-2692页

作者： Hao, Xiaobing Chen, Yuping Ren, Jiadong College of Information Science and Engineering Yanshan University No. 438 Hebei Ave. Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province No. 438 Hebei Ave. Qinhuangdao 066004 China

Sequential pattern mining has a wide range of applications in data streams. The real data involves multiple data streams and each data stream is itemset-sequence. However, most algorithms mine a single item in a single data stream. So IMASP algorithm (incremental mining of across-stream sequential patterns from multiple item set data streams) is presented to mine sequential patterns from multiple data streams and it is an incremental algorithm with single-pass scanning. What is more, the algorithm uses a multi-streams bitmap temporary table to store the sequence information in the current sliding windows. The partial substitution and the real-time counting of the item in old and new windows are adopted to improve the execution efficiency of the algorithm. A multi-streams Lexicographic Sequence Tree is used to store items, extended sequences and the support. Experimental results show that IMASP algorithm reduces time and memory consumption and has the efficiency and accuracy for mining across-stream sequential pattern from itemset sequence streams. © 2014 ICIC International.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Performance analysis of the centralized channel allocation strategy in cognitive radio networks

引用

ICIC Express Letters 2013年第7期7卷 2189-2196页

作者： Dai, Yu Jin, Shunfu Meng, Chao College of Information Science and Engineering Yanshan University No. 438 Hebei Ave Qinhuangdao 066004 China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University No. 438 Hebei Ave Qinhuangdao 066004 China

Based on the centralized channel allocation strategy in the Cognitive Radio Networks, considering the opportunistic channel access for Secondary Users, a discretetime queueing model with a general transmission time and possible transmission interruption is established. By using the method of an embedded Markov chain, the system model is analyzed in steady state, and the performance metrics in terms of the average latency, the throughput for the Secondary User, the channel switching ratio and the channel utilization are given accordingly. Analysis results and simulation experiments are provided to investigate the influence of the traffic load for the Secondary User and the arrival rate for the Primary User packets on the system performance. © 2013 ICIC International.

关键词： Cognitive radio

来源：评论

学校读者我要写书评

暂无评论

A batch-based algorithm for mining weighted sequential patterns in data streams

引用

ICIC Express Letters 2014年第7期8卷 1821-1826页

作者： Hao, Xiaobing Li, Yabin Huang, Guoyan Ren, Jiadong College of Information Science and Engineering Yanshan University No. 438 Hebei Ave. Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province No. 438 Hebei Ave. Qinhuangdao 066004 China

In this paper, we proposed a new sequential pattern mining algorithm called WSPD for mining weighted sequential patterns in data streams. The algorithm produces no false negatives and places a bound on the error of the false positives. WSPD could also satisfy the user's dynamic request for mining weighted sequential patterns from any batches. Experimental results show that WSPD algorithm is effective and scale linearly for mining weighted sequential patterns in data streams. © 2014 ISSN 1881-803X.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

A weighted clustering algorithm based on user’ access sequence similarity

引用

ICIC Express Letters 2016年第10期10卷 2363-2370页

作者： Li, Weina Han, Gaowei Ren, Jiadong College of Information Science and Engineering Yanshan University No. 438 Hebei Ave. Qinhuangdao066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province No. 438 Hebei Ave. Qinhuangdao066004 China

Consider traditional clustering algorithms seldom had a research on users’ visit behavior and content, and they cannot cluster users with similar visit behavior into a community easily. Behavior of user cannot cluster users with similar visit behavior i also reflect usery. Behavior Therefore, setting about the log of user’s access behavior, in view of visitor volume, sequences of accessing keywords and time interval between keywords, a weighted clustering algorithm based on sequence similarity which is called K-Similar is put forward. Starting from behavior of user’s access to document, users are clustered, and important users are divided into a community. Then, literature needs of users in the same cluster are mined. This algorithm has been verified in users users are log data in CNKI, available for academic publishing. Experimental results show that K-Similar is more efficient in efficiency and accuracy. It can obtain more interesting user communities and show academic publishing demand of special users. © 2016 ICIC International.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

An energy saving strategy in digital cognitive radio networks and its performance assessment

引用

ICIC Express Letters 2016年第6期10卷 1367-1373页

作者： Jin, Shunfu Chen, Si School of Information Science and Engineering Yanshan University No. 438 Hebei Ave Qinhuangdao066004 China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University No. 438 Hebei Ave Qinhuangdao066004 China

With the explosive growth of wireless application, how to improve the spectrum efficiency as well as reduce the communication consumption is a hot topic of research. In this paper, we propose a novel energy saving strategy in digital cognitive radio networks (CRNs). The key idea of this strategy is to set the base station (BS) of CRNs into sleep state when there are not any packets to be transmitted. Accordingly, we develop a Markov chain model to assess the average latency of secondary user (SU) packets and the energy saving ratio of system. Finally, we provide numerical results to show that there is a tradeoff when setting the sleep timer length for the proposed energy saving strategy. © 2016 ISSN.

关键词： Energy conservation

来源：评论

学校读者我要写书评

暂无评论

Efficiently mining top-k high utility path patterns in software dynamic call graph

引用

ICIC Express Letters 2016年第1期10卷 211-218页

作者： He, Haitao Yin, Tengteng Wu, Hongfei Ren, Jiadong College of Information Science and Engineering Yanshan University No. 438 West Hebei Ave Qinhuangdao China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University No. 438 West Hebei Ave Qinhuangdao China

Software behavior mining is a very meaningful work. Finding that desirable patterns can assist the program maintainers to comprehend the software adequately. Although the existing high utility pattern mining algorithms can discover all the patterns satisfying a given minimum utility, it is often dicult for users to set a proper minimum utility. In this paper, an ecient algorithm TUPP is presented to mine top-k high utility path patterns in software dynamic call graph. In TUPP, software paths are extracted from software dynamic call graph and the utility of functions is dened rstly. Secondly, a novel structure called PUAList is put forward. The structure stores both the adjacency items and the utility information which is heuristic information for pruning the search space. In addition, two strategies for raising the threshold and one pruning strategy are incorporated into TUPP to increase the eciency. At last, our experiments are conducted on both synthetic and real datasets. The results show that TUPP incorporating the eciency-enhanced strategies demonstrates impressive performance. © 2016 ICIC International.

关键词： Graph theory

来源：评论

学校读者我要写书评

暂无评论

New construction of quaternary sequence with even period based on cyclotomic classes of order two

引用

ICIC Express Letters 2015年第7期9卷 1959-1964页

作者： Jia, Yanguo Shi, Yan Duan, Xiaobei Shen, Xiumin School of Information Science and Engineering Yanshan University No. 438 West Hebei Street Qinhuangdao China The Key Laboratory for Computer Virtual Technology and System Integration of Heibei Province No. 438 West Hebei Street Qinhuangdao China

In this paper, we present a new method for constructing quaternary sequence with even period 2f based on cyclotomic classes of order two. Under the premise of v = 2f +1, we construct two binary sequences with period 2f. Then we can construct a new quaternary sequence from those two binary sequences by using the reverse Gray Mapping. The distribution of autocorrelation function of the proposed quaternary sequence is also derived, which is decided by the parity of f. © 2015, ISSN 1881-803X.

关键词： Binary sequences

来源：评论

学校读者我要写书评

暂无评论

An efficient algorithm for mining maximal frequent sequences by Top-Down Delay Decomposition method

引用

ICIC Express Letters 2014年第7期8卷 2007-2012页

作者： Wang, Qian Ren, Jiadong Liu, Qiang College of Information Science and Engineering Yanshan University No. 438 Hebei Ave. Qinhuangdao 066004 China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province No. 438 Hebei Ave. Qinhuangdao 066004 China

Maximal frequent sequence mining is an important research issue which has realized the highly compressed storage of frequent sequences. At present, most algorithms are based on bottom-up method and large numbers of candidate sequences are generated. In this paper, an algorithm based on Top-down Delay Decomposition for mining Maximal Frequent Sequence is proposed. TDD MFS uses top-down delay decomposition method, sequences from original database and decomposition are expressed by different tuples to record their direct and indirect occurrence counts. Longest infrequent sequences are recursively decomposed and the decomposition of infrequent sequences that have frequent super sequences is avoided. Only one database scan is needed, no candidate sequences are generated and sequence decomposition times are reduced. Experiment result shows its high efficiency and good scalability. © 2014 ISSN 1881-803X.

关键词： Delay decomposition Frequent sequence Maximal frequent sequence

来源：评论

学校读者我要写书评

暂无评论

Improving Vision Transformers with Nested Multi-head Attentions

Improving Vision Transformers with Nested Multi-head Attenti...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Jiquan Peng Chaozhuo Li Yi Zhao Yuting Lin Xiaohan Fang Jibing Gong School of Information Science and Engineering Yanshan University Qinhuangdao China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Microsoft Research Asia Beijing China

Vision transformers have significantly advanced the field of computer vision in recent years. The cornerstone of these transformers is the multi-head attention mechanism, which models interactions between visual elements within a feature map. However, the vanilla multi-head attention paradigm independently learns parameters for each head, which ignores crucial interactions across different attention heads and may result in redundancy and under-utilization of the model’s capacity. To enhance model expressiveness, we propose a novel nested attention mechanism, Ne-Att, that explicitly models cross-head interactions via a hierarchical variational distribution. We conducted extensive experiments on image classification, and the results demonstrate the superiority of Ne-Att.

关键词：

来源：评论

学校读者我要写书评

暂无评论

In-rank: An entropy-based informative nodes mining method in complex software network

引用

ICIC Express Letters 2016年第7期10卷 1683-1689页

作者： Ren, Jiadong Li, Yanling Wu, Hongfei Liu, Yang Zhang, Peng College of Information Science and Engineering Yanshan University No. 438 Hebei Avenue Qinhuangdao066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Yanshan University No. 438 Hebei Avenue Qinhuangdao066004 China

It is a fundamental issue to find a small subset of influential individuals in a complex network such that they can spread information to the largest scope of nodes in the network. Informative functions in complex software network can lead maximum information propagation scope, and mining such function nodes is important to understand the system’s topology structure and information transfer flows. In this paper, a novel topk informative nodes mining approach based on global information in directed-weighted complex software network is proposed. Firstly, we map functions and call relationships between them as a Directed-Weighted Function Call Network (DWFCN). Secondly, an algorithm path-compute based on Software Executing Path Sequence (SEPS) generated by Depth-First-Search strategy is used to compute the software executing paths and get nodes’ information Accessible Set (AS). Thirdly, algorithm Entropy-Compute is proposed to calculate each source node’s information entropy in each information transfer flow process. Finally, an Informative Node Rank (IN-Rank) algorithm is put forward to mine most informative nodes. Experimental results show that our new method is accurate and effective. © 2016, ICIC Express Letters Office. All rights reserved.

关键词： Mining

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：