该文提出了一种基于衰退理论对Flickr数据进行热点事件检测的方法。该方法首先将从Flickr图像中提取的视觉词汇(Visual Words)与图像的文本信息加权合并成文档。然后训练LDA模型获得文档的主题分布作为其最终向量表示。在此基础上提出了一种改进的Single-Pass算法进行事件检测,该算法不仅考虑了图片的地理位置信息,而且基于衰退理论(Aging Theory)对检测到的事件进行生命周期建模,以便计算事件在每个时间段的能量值。最后,根据能量值进行事件排序,获得给定时间段内的热点事件。在真实Flickr数据集上的实验结果表明所提出的方法在精确率、召回率和F1测度上优于传统事件检测方法。
Abstract
This paper proposes an aging theory based method to detect hot events in Flickr data. For each Flickr photo, visual words are first extracted from it and then combined with the content of the attached as a document. An LDA model is trained to predict the topic distribution of each document, which is used as the final vector representation of the document. An improved single-pass clustering algorithm is then proposed to detect events, which take the geographic information of a photo into account. Then aging theory is used to model the life cycles of sequential detected events, determining the energy value of events in each time slot. Finally, hot events in a specific time slot can be detected by ranking the events in terms of their energy value. Experimental results from real Flickr data show that the proposed approach outperforms traditional event detection methods in terms of precision, recall, and F1 value.
Key wordsevent detection; visual words; geographic information; LDA; aging theory
关键词
事件检测 /
视觉词汇 /
地理信息 /
LDA /
衰退理论
{{custom_keyword}} /
Key words
event detection /
visual words /
geographic information /
LDA /
aging theory
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Melanie Aurnhammer, Peter Hanappe, LucSteels. Integrating Collaborative Tagging and Emergent Semantics for Information Retriveal[C]//Proceeding of WWW’06. New York: ACM Press, 2006.
[2] Xirong Li, Cees G M Snoek, Marcel Worring. Learning Tage Relevance by Neighbor Voting for Social Image Retrieval[C]//Proceeding of the 1st ACM International Conference on Multimedia Information Retrieval. New York: ACM Press, 2008: 180-187.
[3] Eva Horter, Rainer Lienhart, Malcolm Slaney. Image Retreival on Large-Scale Image Database[C]//Proceeding of the 6th ACM International Conference Image and Video Retrieval. New York: ACM Press, 2007: 17-24.
[4] J Allan, R Papka, V Lavrenko. On-line new event detection and tracking[C]//Proceedings of the 21st Annual International ACM SIGIR Conference, Melbourne, Australia. ACM Press, 1998: 37-45.
[5] T Brants, F Chen, A Farahat. A System for New Event Detection[C]//Proceedings of the 26th Annual International ACM SIGIR Conference, New York, NY, USA. ACM Press. 2003: 330-337.
[6] H Halpin, V Robu, H Shepherd. The complex dynamics of collaborative tagging[C]//Proceedings of the 16th International Conference on World Wide Web. New York: ACM Press, 2007: 211-220.
[7] Claudiu S Firan, Mihai Georgescu, Wolfgang Nejdl, et al. Bringing Order to Your Photos: Event-Driven Classification of Flickr Images Based on Social knowledge[C]//Proceedings of the 19th ACM International Conference on Information and Knowledge Management. New York: ACM Press, 2010: 189-198.
[8] Hila Becker, Mor Naaman, Luis Gravano. Learning Similarity Metrics for Event Identification in Social Media[C]//Proceedings of 3rd ACM International Conference on Web Search and Data Mining. New York: ACM Press, 2010: 291-300.
[9] Ling Chen, Abhishak Roy. Event Detection from Flickr Data through Wavelet-based Spatial Analysis[C]//Proceedings of the 18th ACM Conference on Information and Knowledge Management. New York: ACM Press, 2009: 523-532.
[10] Xin Jin, Andrew Gallagher, Liangliang Cao, et al. The Wisdom of Social Multimedia: Using Flickr For Prediction and Forecast[C]//Proceedings of the International Conference on Multimedia. New York: ACM Press, 2010:1235-1244.
[11] G P C Fung, J X Yu, H Liu, et al. Time-Dependent Event Hierarchy Construction[C]//Proceedings of KDD2007, California, USA, 2007:300-309.
[12] H L Chieu, Y K Lee. Query Based Event Extraction along a Timeline[C]//Proceedings of the 27th Annual International ACM SIGIR Conference, Sheffield, UK, ACM Press. 2004: 425-432.
[13] Q He, K Chang, E P Lim. Analyzing Feature Trajectories for Event Detection[C]//Proceedings of the 30th Annual International ACM SIGIR Conference, Amsterdam, the Netherlands. ACM Press. 2007: 207-214.
[14] Q He, K Chang, E P Lim. Using Burstiness to Improve Clustering of Topics in News Streams[C]//Proceedings of the 7th IEEE International Conference on Data Mining, 2007: 493-498.
[15] C C Chen, Y T Chen, M C Chen. An Aging Theory for Event Life-Cycle Modeling[J]. IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans, 2007, Vol.37, No.2.
[16] Kuan-Yu Chen, Luesak Luesukprasert, Seng-cho T. Chou. Hot Topic Extraction Based on Timeline Analysis and Multidimensional Sentence Modeling[J]. IEEE Transactions on Knowledge and Data Engineering, IEEE Computer Society, 2007: 1016-1025.
[17] Canhui Wang, Min Zhang, Liyun Ru, et al. Automatic Online News Topic Ranking Using Media Focus and User Attention Based on Aging Theory[C]//Proceeding of CIKM’08, ACM, California, USA. 2008: 1022-1042.
[18] D Blei, A Ng, M Jordan. Latent Dirichlet Allocation[J]. JMLR, 2003, 3:993-1022.
[19] T L Griffiths, M Steyvers. Finding scientific topics[C]//Proceedings of the National Academy of Sciences of the USA. USA: PNAS Press, 2004: 5226-5235.
[20] Bosch A, Munoz X, Marti R. Which is the best way to organize/classiy images by content?[J]. Image and Vision Computing, 2007, 25(6): 778-791.
[21] Lowe DG. Distinctive image features from scale-invariant keypoints[J]. Int’1 Journal of Computer Vision, 2004, 60(2): 91-110.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金资助项目(60970047); 山东省自然科学基金资助项目(Y2008G19);山东大学自主创新基金资助项目(11150070613165)
{{custom_fund}}
PDF(4099 KB)
