正在加载图片...
.342. 智能系统学报 第11卷 的检测,该方法取得了不错的效果。但该模型还有 [10]GAO Jian,DONG Yuwei,SHANG Mingsheng,et al. 待改进,首先,文本的极性判别方式过于粗糙,只是 Group-based ranking method for online rating systems with 考虑了文本中的否定词和关联词,没有特别注重分 spamming attacks[J].EPL (europhysics letters),2015, 110(2):28003. 析程度副词,只分析了文本的极性,却没有定量分析 [11]唐波,陈光,王星雅,等.微博新词发现及情感倾向性 其情感强度;其次,模型中没有动态地考虑评论情 判断分析[J].山东大学学报:理学版,2015,50(1): 况,研究表明,不诚实的商家往往在开办网店的初期 20-25. 雇佣虚假评论人员通过刷单的方式提高自己的信 TANG Bo,CHEN Guang,WANG Xingya,et al.Analysis 誉,所以虚假评论往往发生在电商经营的初期。最 on new word detection and sentiment orientation in Micro- 后,可能评论文本中还隐藏着许多其他的因素可以 blog[J].Journal of Shandong university:nature science, 2015,50(1):20-25. 提高检测的精度,下一步的工作将主要集中在这3 [12]何凤英.基于语义理解的中文博文倾向性分析[J].计 个方面。 算机应用,2011,31(8):2130-2133,2137. 参考文献: HE Fengying.Orientation analysis for Chinese blog text based on semantic comprehension[J].Journal of computer [1]KOLCZ A,ALSPECTOR J.SVM-based filtering of E-mail application,2011,31(8):2130-2133,2137. spam with content specific misclassification costs[C]//Pro- [13]邸鹏,李爱萍,段利国.基于转折句式的文本情感倾向 ceedings of ICDM-2001 Workshop on Text Mining.Dallas, 性分析[J].计算机工程与设计,2014,35(12):4289- USA,2001:324-332. 4295. [2]BECCHETTI L,CASTILLO C,DONATO D,et al.Link- DI Peng,LI Aiping,DUAN Liguo.Text sentiment polarity based characterization and detection of web spam[C]//Ad- analysis based on transition sentence[J].Computer engi- versarial Information Retrieval on the Web.Washington, neering and design,2014,35(12):4289-4295. USA,2006:1012-1021. [14]FENG Song,BANERJEE R,CHOI Y.Syntactic stylometry [3]JINDAL N,LIU Bing.Review spam detection [C]//Pro- for deception detection[C]//Proceedings of the 50th An- ceedings of the 16th International Conference on World nual Meeting of the Association for Computational Linguis- Wide Web.Alberta,Canada,2007:1189-1190. tics:Short Papers-Volume 2.Jeju,Korea,2012:171-175. [4]JINDAL N,LIU Bing,et al.Opinion spam and analysis [15]LI Jiwei,CARDIE C,LI Sujian.TopicSpam:a topic-mod- [C]//Proceedings of the 2008 International Conference on el-based approach for spam detection[C]//Proceedings of Web Search and Data Mining.California,USA,2008:219. the 51st Annual Meeting of the Association for Computa- 230. tional Linguistics.Sofi,Bulgaria,2013:217-221. [5]WU Fang,HUBERMAN B A.Opinion information under [16]JINDAL N,LIU Bing,LIM E P.Finding unusual review costly express[J].ACM transactions on intelligence systems patterns using unexpected rules[C]//Proceedings of the and technology,2010,1(1):5. 19th ACM International Conference on Information and [6]谭文堂,朱洪,葛斌,等.垃圾评论自动过滤方法[J] Knowledge Management.Ontario,Canada,2010:1549- 国防科技大学学报,2012,34(5):153-157,168. 1552. TAN Wentang,ZHU Hong,GE Bin,et al.Method of re- [17]JO Y,OH A H.Aspect and sentiment unification model for view spam detectionJ.Journal of national university of de- online review analysis [C]//Proceedings of the 4th ACM fense technology,2012,34(5):153-157,168. International Conference on Web Search and Data Mining. [7]OTT M.CHOI Y,CARIDIE C,et al.Finding deceptive o- New York,USA,2011:815-824. pinion spam by any stretch of the imagination C//Pro- 作者简介: ceedings of the 49th Annual Meeting of the Association for 赵军,男,1989年生,硕士研究生, Computational Linguistics:human language technologies. 主要研究方向为大数据、数据挖掘、机 Portland,USA,2011,1:309-319. 器学习。 [8]任亚峰,尹兰,姬东鸿.基于语言结构和情感极性的虚 假评论识别J].计算机科学与探索,2014,8(3):313- 320. REN Yafeng,YIN Lan,JI Donghong.Deceptive reviews de- tection based on language structure and sentiment polarity J.Journal of frontiers of computer science and technolo- 王红,女,1966年生,教授,博士生 ,2014,8(3):313-320. 导师,主要研究方向为大数据、复杂网 [9]WANG Guan,XIE Sihong,LIU Bing,et al.Identify online 络数据挖掘。主持国家自然基金项目 store review spammers via social review graph[J].ACM Trans- 1项,参与国家自然基金项目3项,主持 actions on intelligent systems and technology,2012,3(4):61. 省级基金项目6项,发表学术论文 43篇。的检测,该方法取得了不错的效果。 但该模型还有 待改进,首先,文本的极性判别方式过于粗糙,只是 考虑了文本中的否定词和关联词,没有特别注重分 析程度副词,只分析了文本的极性,却没有定量分析 其情感强度;其次,模型中没有动态地考虑评论情 况,研究表明,不诚实的商家往往在开办网店的初期 雇佣虚假评论人员通过刷单的方式提高自己的信 誉,所以虚假评论往往发生在电商经营的初期。 最 后,可能评论文本中还隐藏着许多其他的因素可以 提高检测的精度,下一步的工作将主要集中在这 3 个方面。 参考文献: [1] KOLCZ A, ALSPECTOR J. SVM⁃based filtering of E⁃mail spam with content specific misclassification costs[C] / / Pro⁃ ceedings of ICDM⁃2001 Workshop on Text Mining. Dallas, USA, 2001: 324⁃332. [2] BECCHETTI L, CASTILLO C, DONATO D, et al. Link⁃ based characterization and detection of web spam[C] / / Ad⁃ versarial Information Retrieval on the Web. Washington, USA, 2006: 1012⁃1021. [3] JINDAL N, LIU Bing. Review spam detection [ C] / / Pro⁃ ceedings of the 16th International Conference on World Wide Web. Alberta, Canada, 2007: 1189⁃1190. [4] JINDAL N, LIU Bing, et al. Opinion spam and analysis [C] / / Proceedings of the 2008 International Conference on Web Search and Data Mining. California, USA, 2008: 219⁃ 230. [5] WU Fang, HUBERMAN B A. Opinion information under costly express[J]. ACM transactions on intelligence systems and technology, 2010, 1(1): 5. [6]谭文堂, 朱洪, 葛斌, 等. 垃圾评论自动过滤方法[ J]. 国防科技大学学报, 2012, 34(5): 153⁃157, 168. TAN Wentang, ZHU Hong, GE Bin, et al. Method of re⁃ view spam detection[J]. Journal of national university of de⁃ fense technology, 2012, 34(5): 153⁃157, 168. [7]OTT M, CHOI Y, CARIDIE C, et al. Finding deceptive o⁃ pinion spam by any stretch of the imagination [ C] / / Pro⁃ ceedings of the 49th Annual Meeting of the Association for Computational Linguistics: human language technologies. Portland, USA, 2011, 1: 309⁃319. [8]任亚峰, 尹兰, 姬东鸿. 基于语言结构和情感极性的虚 假评论识别[J]. 计算机科学与探索, 2014, 8(3): 313⁃ 320. REN Yafeng, YIN Lan, JI Donghong. Deceptive reviews de⁃ tection based on language structure and sentiment polarity [J]. Journal of frontiers of computer science and technolo⁃ gy, 2014, 8(3): 313⁃320. [9] WANG Guan, XIE Sihong, LIU Bing, et al. Identify online store review spammers via social review graph[J]. ACM Trans⁃ actions on intelligent systems and technology, 2012, 3(4): 61. [10 ] GAO Jian, DONG Yuwei, SHANG Mingsheng, et al. Group⁃based ranking method for online rating systems with spamming attacks[J]. EPL (europhysics letters), 2015, 110(2): 28003. [11]唐波, 陈光, 王星雅, 等. 微博新词发现及情感倾向性 判断分析[ J]. 山东大学学报:理学版, 2015, 50( 1): 20⁃25. TANG Bo, CHEN Guang, WANG Xingya, et al. Analysis on new word detection and sentiment orientation in Micro⁃ blog[J]. Journal of Shandong university: nature science, 2015, 50(1): 20⁃25. [12]何凤英. 基于语义理解的中文博文倾向性分析[ J]. 计 算机应用, 2011, 31(8): 2130⁃2133, 2137. HE Fengying. Orientation analysis for Chinese blog text based on semantic comprehension[J]. Journal of computer application, 2011, 31(8): 2130⁃2133, 2137. [13]邸鹏, 李爱萍, 段利国. 基于转折句式的文本情感倾向 性分析[J]. 计算机工程与设计, 2014, 35(12): 4289⁃ 4295. DI Peng, LI Aiping, DUAN Liguo. Text sentiment polarity analysis based on transition sentence[ J]. Computer engi⁃ neering and design, 2014, 35(12): 4289⁃4295. [14]FENG Song, BANERJEE R, CHOI Y. Syntactic stylometry for deception detection[C] / / Proceedings of the 50th An⁃ nual Meeting of the Association for Computational Linguis⁃ tics: Short Papers⁃Volume 2. Jeju, Korea, 2012: 171⁃175. [15]LI Jiwei, CARDIE C, LI Sujian. TopicSpam: a topic⁃mod⁃ el⁃based approach for spam detection[C] / / Proceedings of the 51st Annual Meeting of the Association for Computa⁃ tional Linguistics. Sofi, Bulgaria, 2013: 217⁃221. [16] JINDAL N, LIU Bing, LIM E P. Finding unusual review patterns using unexpected rules [ C] / / Proceedings of the 19th ACM International Conference on Information and Knowledge Management. Ontario, Canada, 2010: 1549⁃ 1552. [ 17]JO Y, OH A H. Aspect and sentiment unification model for online review analysis[C] / / Proceedings of the 4th ACM International Conference on Web Search and Data Mining. New York, USA, 2011: 815⁃824. 作者简介: 赵军,男,1989 年生, 硕士研究生, 主要研究方向为大数据、数据挖掘、机 器学习。 王红,女,1966 年生,教授,博士生 导师,主要研究方向为大数据、复杂网 络、数据挖掘。 主持国家自然基金项目 1 项,参与国家自然基金项目 3 项,主持 省级 基 金 项 目 6 项, 发 表 学 术 论 文 43 篇。 ·342· 智 能 系 统 学 报 第 11 卷
<<向上翻页
©2008-现在 cucdc.com 高等教育资讯网 版权所有