1SSN1673-4785 CN23-1538/TP AA 中国人工智能学会会刊 智能系统学报 CAAI TRANSACTIONS ON INTELLIGENT SYSTEMS 登三系缠学招 7oASartos IssN1673-4785 4 08> 2012 771673478120 Vol.7 No.4
《智能系统学报》第二届编辑委员会委员名单 The Second Editorial Board of CAAI Transactions on Intelligent Systems 顾 问Consultants 杨子YANG Shuzi 李行达LI Yanda 吴启迪WUQd 张钱ZHANG Bo 郑南宁ZHENG Nannin吨 赵沁平ZHA0 Qinping涂序查TU Xuyan 在保宗YUAN Baong 童天湘TONG Tianxiang 董祖美DONG Yunm 主 任Chairman何新贵HE Xingui 副主 任Vice Chairmen 徐玉如XU Yun 蔡自兴CAI Zixing 孙增圻SUN Zen可 秦世引QIN Shiyin 王科俊WANG Kejun 委 员Members 丁水生DING Yongsheng 马少平MA Shaoping 马世龙MA Shilong 尹怡欣YIN Yiin 王龙WANG Long 王万森WANG Wansen 王飞跃WANGFeiyue 王立权WANG Liqu 王志良WANG Zhiliang 王国胤WANG Guoyin 王科復WANG Kejun 冯嘉礼FENGJiali 史忠植SHⅢZhongzhi 田金文TIAN Jinwen 朱齐丹ZHU Qidan 庄越挺ZHUANG Yueting 刘丁uUDi 刘民UM 刘宏LJU Hong 刘胜LIU Sheng 刘清UQi吨 刘增良LIU Zengliang 孙增折SUN Zengri 配秋骑RUAN Qiup 何清HE Qing 何华灿HE Huacan 何新贵H亚Xingi 吴快军WU Tiejun 张军ZHANG Jun 张长水ZHANG Changhui 张汝被ZHANG Rubo 张铭钩ZHANG Mingjun 杜军平D0 Junping 迟惠生C印Huisheng 邱玉辉QIU Yuhui 陈杰CHEN Jie 范多旺FAN Duowan唱 周志华ZHOU Zhih 施要飞SHI Pengfei 春红卷ZHA Hongbin 段海滨DUAN Haibin 赵琳ZHAO Lin 钟义信ZHONG Yigin 除玉如XUYuru 徐若冰XU Ruobing 秦世引QIN Shiyi面 贾英民LA Yingmi血 郭军GU0Jun 黄心汉HUANG Xinhar 曹元大CAO Yunda 焦李成IAO Lichen% 韩力群HAN Liqun 鲁华样LU Huxinng 蔡文CAI Wen 蔡自兴CAI Zixing 国间International Consultants G.Sanniti Di Baja,Istituto di Cibernetica,CNR.Italy T.J.Tamn.Washington University,USA 国际委员International Members Don Hong.Middle Temesee Stste University,USA Fanika Mercier-Laurent,Jean Moulin Univenity Lymn 3,French Hyouk Ryeol Choi,Sungkyunkwan University.Koree Jean-Claude Latomhe,Stanford University,USA Jiebo Luo.Eastmun Kodak Company.USA Technology,Japun Laurent Itti.University of Southem Califomia,USA Randall Davis.Maseschu setts Institute of Technology,USA Reinhard Klette,The University of Auckland,New Zealand Ronald L.Yager,Mohammed VUniversity,Moroo Shuzhi Sam Ge,The National Univerity of Singpore,Singpore Wankyun Chung,POSTECH,Koree Xindong Wu.University of Vermiont,USA 编Chief Editor 钟义信ZHONG Yixin 副 编Vice Chief Editor 徐若冰XU Ruobing 编辑部主任Director 徐若冰XU Ruobing 编辑部副主任Deputy Directors 刘玉明LIU Yuming 李雪莲L山Xuelian 责任编辑Responsible Editors 马兰兰MA Lanle 刘充充IU Liangliang 英文编兼)English Editor(Part-time) 朱玉珍ZHU Ywhen Andrew Knox (USA)
智能系统学报 第7卷第4期 (总第36期) 2012年8月 目 次 基于外骨骼机器人技术的人体手臂震颤抑制的理论和方法 孙建,向馗,高理富,李涛,葛运建(283) 相关向量机分类方法的研究进展与分析 …赵春晖,张燚(294) 基于PGA的全流水双精度浮点矩阵乘法器设计 4… 刘沛华,鲁华祥,龚国良,刘文鹏(302) 基于文本的新闻事件多版本发现模型 …肖融,孔亮,张岩(307) 图像复原中的模糊参数估计 王伟,郑津津,刘星,周洪军,沈连缩(315) 基于止血机制的冗余并联机器人精准容错控制 郭崇滨,郝矿荣,丁永生(321) 基于关联词的主题模型语义标注 … 周亦鹏,杜军平(327) 视觉关注转移的事件检测算法 …张丽坤,孙建德,李静(333) 脑电信号的小波变换和样本嫡特征提取方法 ……张毅,罗明伟,罗元(339) 自适应扩维UKF算法在SINS/GPS组合导航系统中的应用 孙尧,马涛,高延滨,王璐(345) 离散时间混合多智能体的拟平均一致性控制 李波,吴淑琴,谷明琴(352) 一种虚拟人追逐过程中的情绪博弈模型 卞玉龙,刘箴(358) 面向学术社区的专家推荐模型 ........... 李春荚,汤庸,陈国华,汤志康(365) 混合改进蚁群算法的函数优化 陈明杰,黄佰川,张旻(370) ·信息交流· 第4届群体智能国际会议征文通知 0年0。。00.0.00。。。00。g00。。0000年000.0000卡。。0+0”00.0。。 (314) 第9届国际机器学习和数据挖掘会议 (338) 2012年第2届计算机科学与网络技术国际会议 ............................................... (344) 第四届全国智能信息处理学术会议(NCP2013)征文通知 (351) 2012年自动化控制和机器人技术国际会议 (364) 2013年EEE机器人和自动化国际会议(ICRA2013) (376) 期刊基本参数:CN23-1538/TP*2006*b*A4*94*zh*P*¥15.00*1500*14*2012-08
CAAI Transactions on Intelligent Systems Vol.7 No.4 (Sum No.36) Aug.2012 Contents A comprehensive review of fundamental theory and methodology for tremor suppression of human arm based on robotic exoskeleton technology..SUN Jian,XIANG Kui,GAO Lifu,LI Tao,GE Yunjian(283) Research progress and analysis on methods for classification of RVM ............................................HAO chunhui,ZHANG Vi(294) Design of an FPGA-based double-precision floating-point matrix multiplier with pipeline architecture ........LIU Peihua,LU Huaxiang,GONG Guoliang,LIU Wenpeng (302) A text clustering model for diverse versions discovery ................XIAO Rong,KONG Liang,ZHANG Yan(307) Estimation of blur parameters in image restoration .......WANG Wei,ZHENG Jinjin,LIU Xing,ZHOU Hongjun,SHEN Lianguan(315) Hemostasis mechanism based precise fault-tolerant control for redundant parallel manipulator ........GUO Chongbin,HAO Kuangrong,DING Yongsheng(321) Semantic tagging of a topic model based on associated words …………e…… ZHOU Yipeng,DU Junping(327) Event detection based on visual attention shift …ZHANG Likun,SUN Jiande,LI Jing(333)) EEG feature extraction method based on wavelet transform and sample entropy …ZHANG Yi,.LU0 Mingwei,LU0Yuan(339) An adaptive augmented unscented Kalman filter with applications in a SINS/GPS integrated navigation system .....................SUN Yao,MA Tao,GAO Yanbin,WANG Lu(345) Quasi-average-consensus control of hybrid swarm agents with discrete time ..............................................................................LIBo,wU Shugin,cU Minggin(352) A model of emotion game for chasing between virtual humans … BIAN Yulong,LIU Zhen (358) Research on an expert recommendation model based on the scholar community SCHOLAT ...........LI Chunying,TANG Yong,CHEN Guohua,TANG Zhikang(365) Function optimization based on an improved hybrid ACO CHEN Mingjie,HUANG Baichuan,ZHANG Min(370) International The Fourth International Conference on Swarm Intelligence (ICSI'2013)..........................(314) 9th International Conference on Machine Learning and Data Mining (MLDM)2013..........................(338) 2nd International Conference on Computer Science and Network Technology(ICCSNT 2012)...............(344) The 4th National Conference on Intelligent Information Processing.......(351) International Conference on Automation,Control and Robotics (ICACR'2012).........(364) The 2013 IEEE International Conference on Robotics and Automation (ICRA 2013)..................(376) Serial parameters:CN23-1538/TP*2006*b*A4*94*zh*P*¥15.00*1500*14*2012-08
第7卷第4期 智能系统学报 Vol.7 No.4 2012年8月 CAAI Transactions on Intelligent Systems Aug.2012 D0I:10.3969/i.issn.16734785.201204017 网络出版t地址:htp://www.cnki.net/kcma/detail/23.1538.TP.20120712.1111.008.html 基于关联词的主题模型语义标注 周亦鹏,杜军平2 (1.北京工商大学计算机与信息工程学院,北京100048;2.北京邮电大学智能通信软件与多蝶体北京市重,点实验 室,北京100876) 摘要:互联网主题分析中经常采用概率主题模型对主题进行描述,但存在对于一般用户难以理解的问题,提出一 种概率主题模型的自动语义标注方法.首先通过基于语义分类的关联规则挖掘关联主题词并建立候选标签集合,然 后以关联词在数据集中的概率分布来设计相关性判别函数,计算候选标签和主题模型的相关度,最后根据最大边缘 相关选择高语义覆盖度和区分度的标签.在食品安全和旅游领域主题模型标注的实验表明,与最大概率主题词标记 方法相比,提出的方法能够明显提高标注的准确性,并且解决了多标签标记中语义类别单一的问题,能够以较少数 量的标签表达更为丰富的语义,这有助于进一步实现更为准确的主题跟踪和主题信息检索。 关键词:主题分析;语义标注;生成模型;关联词;关联规则 中图分类号:TP391文献标志码:A文章编号:16734785(2012)04032706 Semantic tagging of a topic model based on associated words ZHOU Yipeng,DU Junping (1.School of Computer Science and Information Engineering,Beijing Technology and Business University,Beijing 100048,China; 2.Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia,Beijing University of Posts and Telecommunica- tions,Beijing 100876,China) Abstract:In topic analysis field of Internet,the probabilistic topic model is often used to describe topic seman- teme.But the semanteme of a topic model is difficult for users to understand.An automatic semantic tagging meth- od of a probabilistic topic model is proposed.Firstly,an association rule mining algorithm based on semantic cate- gories is presented to get associated topic words,which consist of a candidate tag set.Then,according to the prob- ability of associated words,a semantic correlation function is used to calculate semantic correlation of candidate tags and topic model.At last,a maximal marginal relevance method is used to select tags with better semantic coverage and discrimination.The experimental results of food safety and tourism topic model proved that,compared with maximum probability topic words tagging method,the proposed method can improve accuracy of topic tagging obvi- ously,and can express more abundant semantemes with a small number of tags,which solve the problem of single semantic category in the multi-tagging method.So it is helpful to achieve more accurate topic tracking and topic in- formation retrieval. Keywords:topic analysis;semantic tagging;generative model;associated words;association rule 主题分析通常采用概率生成模型,如LDA、PL是取概率较高的若干个语义词来表示主题含义3], SA等方法,以语义词概率分布的形式描述主题12], 但这种方法也常常不能准确表示整个分布所覆盖的 这使得一般用户较难理解主题的内容.通常的方法 全部语义.因此,提出一种主题模型的自动标注方 法,提取具有一定语义覆盖度和区分度的主题关联 收稿日期:2012-04-20.网络出版日期:201207-12. 词来描述主题的内容。 基金项目:国家“973”计划资助项目(2012CB821206):国家自然科学 基金资助项目(91024001,61070142);北京市自然科学基 1主题模型的语义标注 金资助项目(4111002). 通信作者:周亦鹏.Email:ripengzhou(@l63.com 主题模型的自动语义标注通常包括2个步骤: