Senior Design Presentation
Transcript: Integrated Correlation Factor (ICF) *Issue: Pearson correlation doesn't exist some times *Address: assign default value MCA Acknowledgment *MCA is an extension of the standard correspondence analysis to more than two variables *It demonstrates the robustness and relatively high accuracy in modeling the posterior probability distribution Positive and negative Overcome semantic gap ICF-Result rectification *Full name: Pearson product-moment correlation coefficient *Our negative correlation selection result is better than the existing algorithms Yang Liu Advisor: Dr. Mei-Ling Shyu References Cont Introduction MCA(Multiple correspondence Analysis) Positive and negative *Problem: How to combine the scores from all the components? *Solution: Logistic regression and optimization *Objective: To address the bias data issue Introduction: multimedia data *Minimize cost function: Constrained convex optimization problem because * Gradient descent method • Tao Meng,Yang Liu, Mei-Ling Shyu, Yilin Yan and Chi-Min Shu, “Enhancing Multimedia Concept Mining and Retrieval by Incorporating Negative Correlations”, IEEE International Conference on Semantic Computing (IEEE-ICSC) (Accepted) • Yilin Yan, Yang Liu and Mei-Ling Shyu, “Imbalanced data classification using concept correlation”, IEEE international conference on Information Reuse and Integration (IRI) (Under review) [1] E. Gabarron, L. Fernandez-Luque, M. Armayones, and A. Y. Lau,“Identifying measures used for assessing quality of youtube videos with patient health information: A review of current literature,” Interactive Journal of Medical Research, vol. 2, no. 1, March 2013. [2] T. Meng and M.-L. Shyu, “Automatic annotation of drosophila developmental stages using association classification and information integration,” in The 12th IEEE International Conference on Information Resue and Integration (IRI 2011), Las Vegas, Nevada, August 2011, pp. 142–147. [3] M.-L. Shyu, Z. Xie, M. Chen, and S.-C. Chen, “Video semantic event/concept detection using a subspace-based multimedia data mining framework,” IEEE Transactions on Multimedia, vol. 10, pp. 252–259, February 2008. [4] L. Lin, M.-L. Shyu, and S.-C. Chen, “Correlation-based interestingness measure for video semantic concept detection,” in 2009 IEEEInternational Conference on Information Reuse & Integration, 2009, pp. 120–125. [5] Q. Zhu, L. Lin, M.-L. Shyu, and S.-C. Chen, “Feature selection using correlation and reliability based scoring metric for video semantic detection,” in 2010 IEEE Fourth International Conference on Semantic Computing (ICSC), 2010, pp. 462–469. [6] A. F. Smeaton, P. Over, and W. Kraaij, “Evaluation campaigns and TRECVid,” in Proceedings of the 8th ACM International Workshop on Multimedia Information Retrieval, October 2006, pp. 321–330. [7] E. A. Cherman, J. Metz, and M. C. Monard, “Incorporating label dependency into the binary relevance framework for multi-label classification,” Expert Systems with Applications, vol. 39, no. 2, pp. 1647–1655, February 2011. [8] J. R. Smith, M. Naphade, and A. Natsev, “Multimedia semantic indexing using model vectors,” in IEEE International Conference on Multimedia and Expo, Baltimore, MD, June 2003, pp. 445–448. [9] M. R. Naphade, I. Kozinetsey, T. S. Huang, and K. Ramchandran, “A factor graph framework for semantic indexing and retrieval in video,” in the IEEE Workshop on Content-based Access of Image and Video Libraries, Washington, DC, June 2000, pp. 35–39. [10] T. Meng and M.-L. Shyu, “Leveraging concept association network for multimedia rare concept mining and retrieval,” in IEEE International Conference on Multimedia and Expo, Melbourne, Australia, July 2012. [11] L. Ballan, M. Bertinti, A. D. Bimbo, and G. Serra, “Video annotation and retrieval using ontologies an rule learning,” IEEE Multimedia, vol. 17, no. 4, pp. 80–88, October-December 2010. Thank You Score Integration cont:Logistic regression Equation Score Integration * Multimedia Information Retrieval (MMIR or MIR) is a research discipline of computer science that aims at extracting information from multimedia data sources. * Semantic gap * The gap between low level features and high level concepts of images. What is multimedia retrieval? How can inter concept correlations contribute to this retrieval process? Introduction *Our integration steps can be considered as two layer perceptron （Neural network） Final equation ICF-Pearson Correlation My parents Lianzhan Liu and Xiujun Yao Data mining, Database & Multimedia Research Group * Dr.Shyu * Tao Meng, Dianting Liu, Qiusha Zhu, YiLin Yan Thank you for the faculty members of ECE department Score Integration cont:Optimization Accomplishment * Positive: *Concepts that help increase the possibility of co-occurrence * Negative *Concepts that help decrease the possibility of co-occurrence *Positive: A lot of work have been done *Smith, J.R., Naphade, M., Natsev, A.: Multimedia semantic indexing using model vectors. *Meng, T., Shyu, M.L.: Leveraging
Prezi Present
Prezi Video
Prezi Design
