检索结果-内蒙古大学图书馆

Communications in computer and Information Science 2014年 480卷 156-166页

作者： Zhou, Jian Wang, Xin Feng, Zhiyong School of Computer Science and Technology Tianjin University Tianjin China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin China

With the continuous growth of ontologies in size, the amount of ontology axioms is steadily increasing. To help users understand ontologies with a large number of axioms, we devise a novel ontology visualization tool, called AxiomVis, which defines a mapping from DL axioms to the directed acyclic graph (DAG). We give a recursive procedure to generate the target DAG and then propose a new axiom-aware layout algorithm to display the DAG elements using the characteristics of the expression tree. The distinguishing feature of our tool is that it visualizes ontologies from the axiom viewpoint. This paper demonstrates comparison of different layouts, which exhibits the advantages of our layout on ontology visualization. © Springer-Verlag Berlin Heidelberg 2014.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Lable-based hybrid recommendation algorithm: Implementation and analysis

引用

Journal of Information and Computational Science 2015年第6期12卷 2221-2228页

作者： Yu, Ruiguo Mao, Hongyue Wang, Zan Zhang, Minjie Yuan, Zhenxia He, Muwen Chang, Peng School of Computer Science and Technology Tianjin University Tianjin China School of Computer and Software Tianjin University Tianjin China Information and Network Center Tianjin University Tianjin China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin China

Recommendation system, as an information filtering method, has been widely applied and the hybrid recommendation algorithm is by far the most commonly used technique. But in the field of online learning system, recommendation system is not mature enough. In this paper, we analyzed the Tianjin University Online Judge, established a model to apply the recommendation algorithm to it, and then proposed a label-based hybrid recommendation algorithm which is applicable to the system. Also, by conducting experiments on real users' data of Tianjin University Online Judge System, we evaluated our algorithm and obtained good results. Copyright © 2015 Binary Information Press.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Research on personalized recommendation algorithm based on user’s attention behavior

引用

ICIC Express Letters 2016年第11期10卷 2615-2622页

作者： Yu, Mei Di, Haibo Hou, Dejun Ma, Huijuan Pei, Yingjie Li, Zengjie Tianjin Key Laboratory of Cognitive Computing and Application No. 92 Weijin Road Nankai Dist Tianjin300072 China School of Computer Science and Technology Management Tianjin University No. 92 Weijin Road Nankai Dist Tianjin300072 China School of Computer Software Management Tianjin University No. 92 Weijin Road Nankai Dist Tianjin300072 China Office of State-Owned Assets and Equipment Management Tianjin University No. 92 Weijin Road Nankai Dist Tianjin300072 China Karamay Power Supply Company Xinjiang Oilfield Company Petro China Karamay834000 China The Thermal Power Plant of Dushanzi Petrochemical Company Karamay833699 China

Nowadays, how to deal with mass data and to perform personalized recommendation for users has become a research hotspot. Collaborative filtering recommendation algorithm has been widely used since it was proposed. However, there is still more room for improvement in the computation of similarity. This paper introduces user’s attention behavior on the basis of the traditional homogeneity measure, and puts forward a personalized recommendation algorithm based on user’s attention behavior which combines with the low rank matrix factorization technique to improve the accuracy of the predictions. We named the algorithm pTrust. Through contrast experiments on a public dataset, the pTrust algorithm shows better effectiveness both on accuracy and time complexity in the trust network. © 2016 ICIC International.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

3D garment segmentation based on semi-supervised learning method

引用

Journal of Fiber Bioengineering and Informatics 2015年第4期8卷 657-665页

作者： Huang, Mian Liu, Li Liu, Ruomei Fu, Xiaodong Liu, Lijun Yunnan Provincial Key Laboratory of Computer Technology Application Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China National Engineering Research Center of Digital Life State-Province Joint Laboratory of Digital Home Interactive Applications School of Data and Computer Science Sun Yat-sen University Guangzhou China

In this paper, we propose a semi-supervised learning method to simultaneous segmentation and labeling of parts in 3D garments. The key idea in this work is to analyze 3D garments using semi-supervised learning method which can label parts in various 3D garments. We first develop an objective function based on Conditional Random Field (CRF) model to learn the prior knowledge of garment components from a set of training examples. Then, we exploit an effective training method that utilizes JointBoost classifiers based on the co-analysis for garments. And we modify the JointBoost to automatically cluster the segmented components without requiring manual parameter tuning. The purpose of our method is to relieve the manual segmentation and labeling of components in 3D garment collections. Finally, the experimental results show the performance of our proposed method is effective. © 2015 Binary Information Press & Textile Bioengineering and Informatics Society December 2015.

关键词： Random processes

来源：评论

学校读者我要写书评

暂无评论

A survey of major techniques of clothing in vision

引用

Journal of Fiber Bioengineering and Informatics 2015年第4期8卷 725-732页

作者： Huang, Dongyan Liu, Li Fu, Xiaodong Liu, Lijun Huang, Qingsong Yunnan Provincial Key Laboratory of Computer Technology Application Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China National Engineering Research Center of Digital Life State-Province Joint Laboratory of Digital Home Interactive Applications School of Data and Computer Science Sun Yat-sen University Guangzhou China

Clothing in vision plays an important role in many applications of digital clothing, virtual fitting, fashion design and manufacturing. The research of clothing in vision is full of technical challenges, such as background detection and features extraction. In this paper, we introduce the latest remarkable works of clothing in vision area, including clothing segmentation, clothing recognition and clothing retrieval. There are many remarkable advances on the basis of the predecessors' achievements in the three fields. Firstly, we provide an overview of the techniques for clothing segmentation and clothing recognition. Then, we introduce and analyze the representative works on clothing retrieval. Finally, we forecast broader application prospect and future research work of clothing in vision. © 2015 Binary Information Press & Textile Bioengineering and Informatics Society December 2015.

关键词： Textiles

来源：评论

学校读者我要写书评

暂无评论

An empirical study of phonetic transfer in English monophthong learning by Tibetan (Lhasa) speakers 18

An empirical study of phonetic transfer in English monophtho...

引用

18th Annual International Conference Oriental COCOSDA, O-COCOSDA 2015 - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, CASLRE 2015

作者： Hui, Feng Lu, Zhao Jianwu, Dang School of Foreign Languages and Literature Tianjin University Tianjin China Tianjin Key Laboratory of Cognitive Computing and Application Tianjin China Research Center for Linguistic Sciences Tianjin University Tianjin China Department of Publicity and Counselling Tianjin Library Tianjin China School of Computer Science and Technology Tianjin University Tianjin China School of Information Science JAIST Japan

ISBN: (纸本)9781467382793

English teaching has been carried out in every region of China. However, the teaching of English phonetics in minority regions faces the problems of poor teaching quality and low intelligibility of phonetic production. Under the framework of Speech Learning Model (SLM), the current research adopts the experimental phonetic approach to select 10 Lhasa students (5 male and 5 female) from the Corpus of Chinese, English, Tibetan by Tibetan Speakers (CETTS) and examine the characteristics of the vowel space of their English in continuous speech, with a purpose to explore minority students' phonetic transfer of L1 and L2 on L3 vowel system. Euclidean distances between Tibetan speakers' English and Tibetan, and between Tibetan speakers' English and Chinese, reveal the degree of phonetic transfer of Tibetan (L1) and Chinese (L2) on English (L3). Results show that the nonstandard English vowel production by Lhasa speakers are more influenced by the similar vowels in L1 or L2. Due to equivalence classification, Lhasa students' English vowels /a:/, /i:/ and /i/ are more influenced by Tibetan similar vowels, and their English vowels /ω/ and /u:/ are more influenced by Chinese similar vowels. In addition, Tibetan female speakers' / is more influenced by the similar vowel in Chinese. The study has revealed that when Tibetan speakers learn L3, the interference seems to be partly from their L1 and partly from their L2. Therefore, in minority students' L3 vowel learning, both L1 and L2 have influence on their L3 vowel system. © 2015 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

System of automatic Chinese webpage summarization based on the random walk algorithm of dynamic programming

引用

Open Cybernetics and Systemics Journal 2015年第1期9卷 1315-1322页

作者： Wang, Feng Qin, Xiaoming Wang, Yizhen Wei, Xinjiang School of Mathermatics and Statistics Science Ludong University Yantai264025 China Key Laboratory of Language Resource Development and Application of Shandong Province Yantai264025 China College of Computer and Information Engineering Jiaozuo Teachers College Jiaozuo454000 China Institute of Computer Application Technology Jiaozuo Teachers College Jiaozuo454000 China Tilburg University Tilburg Netherlands

As the Internet becomes more and more deeply connected with our life, the Internet has brought together mass text material, and it is still in explosive growth. In order to quickly and accurately to help users find the required content, the traditional solution is to use a search engine. However, the results of existing automatic webpage summarization systems for search engine are of low quality. Because they just based on statistical method, gather some sentences in the web document beside the search phrases. Neither symbolizes the subject of the document, nor take into account the user search phrases. According to the shortages, An automatic webpage summarization systems is realized. On the basis of the work done, this paper proposed an automatic text summarization method based on relation graph and text structure analysis. This method firstly segment text into semantic paragraphs. For each semantic paragraph, a subject term discover method based on relation graph analysis is proposed. At last, both search phrase and document subject are take into account, it extracts summary according to the guidance of the subject terms. © Wang et al.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Identification of emission sources of umbral flashes using phase congruency

引用

Research in Astronomy and Astrophysics 2014年第8期14卷 1001-1010页

作者： Song Feng Lan Yu Yun-Fei Yang Kai-Fan Ji Yunnan Key Laboratory of Computer Technology Application/Faculty of Information Engineering and Automation Kunming University of Science and Technology Key Laboratory of Solar Activity National Astronomical ObservatoriesChinese Academy of Sciences Department of Information Engineering Yunnan Land and Resources Vocational College

The emission sources of umbral flashes （UFs） are believed to be closely related to running umbral and penumbral waves, and are concluded to be associated with umbral dots in the solar photosphere. Accurate identification of emission sources of UFs is crucial for investigating these physical phenomena and their inherent relationships. A relatively novel model of shape perception, namely phase congruency （PC）, uses phase information in the Fourier domain to identify the geometrical shape of the region of interest in different intensity levels, rather than intensity or gradient. Previous studies indicate that the model is suitable for identifying features with low contrast and low luminance. In the present paper, we applied the PC model to identify the emission sources of UFs and to locate their positions. For illustrating the high performance of our proposed method, two time sequences of Ca n H images derived from the Hinode/SOT on 2010 August 10 and 2013 August 20 were used. Furthermore, we also compared these results with the analysis results that are identified by the traditional/classical identification methods, including the gray-scale adjusted technique and the running difference technique. The result of our analysis demonstrates that our proposed method is more accurate and effective than the traditional identification methods when applied to identifying the emission sources of UFs and to locating their positions.

关键词： techniques image processing -- Sun chromosphere -- Sun photo- sphere -- sunspots -- convection

来源：评论

学校读者我要写书评

暂无评论

Effects of early afterdepolarizations on ventricular tachycardia in human heart

Effects of early afterdepolarizations on ventricular tachyca...

引用

computers in Cardiology (CinC)

作者： Jieyun Bai Kuanquan Wang Qince Li Yinghui Li Henggui Zhang School of Computer Science and Technology Harbin Institute Technology Harbin China Harbin Institute of Technology Harbin China State Key Laboratory of Space Medicine Fundamentals and Application Beijing China School of Physics and Astronomy University of Manchester Manchester UK

ISBN: (纸本)9781509006601

Experimental studies suggest EADs may occur at rapid heart rates as a consequence of tachyarrhythmias. The aim of this study was to investigate the interaction between EAD and rapid reentrant excitation waves and assess its effects with electrocardiogram (ECG).The simulation results indicated that, at the cellular level, reduced repolarization reserve contributed to action potential duration (APD) prolongation (ENDO: 302ms vs. 402ms, MIDDLE: 414ms vs.>1000ms, EPI: 298ms vs. 397ms) and genesis of EADs only in MIDDLE cells. In the 3D model, EADs caused drift of rapid rotors. Multiple focal excitations arising from EADs kept regeneration of reentrant excitation waves by breaking excitation wave fronts. ECGs presented periodic features with stable reentry in control condition, but degenerated into irregular and complex patterns in EADs condition. The simulation results demonstrate that MIDDLE cells are prone to genesis of EADs at rapid heart rates, which plays an important role in degenerating ventricular tachycardia into ventricular fibrillation.

关键词： Numerical models Electric potential Lead Electrocardiography Computational modeling Heart Three-dimensional displays

来源：评论

学校读者我要写书评

暂无评论

Reliability Analysis Based on Coupled Relationship

Reliability Analysis Based on Coupled Relationship

引用

2015 International Conference on Artificial Intelligence and Industrial Engineering(AIIE 2015)

作者： W.L.Pan H.Wang School of Mathematics & Computer Science Yunnan Minzu University Key Laboratory of IOT Application Technology of Universities in Yunnan Province Yunnan Minzu University College of Mechanical and Electrical Engineering Harbin Engineering University

ISBN: (纸本)9781510806450

In the usual reliability analysis, researchers often acquire the reliability information about a product by reliability tests and usually regard the product as a whole object to study. Also, the relationships between the attributes of the product are usually ignored. Base on the non-IIDness learning, the attributes of objects are more or less interacted and coupled. Thus, this paper analyzes the coupled relationships between each pair of product attributes and the coupled relationships between each product attribute and reliability lifetime of the product. Specifically, the attributes we analyzed are continuous. The level of interplay between each pair of attributes and between each attribute and product lifetime can be observed by coupled correlation matrices.

关键词： reliability non-IIDness learning coupled relationship

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：