Multimodal non-linear latent semantic method for information retrieval

dc.contributor.advisorGonzález Osorio, Fabio Augustospa
dc.contributor.advisorVanegas Ramírez, Jorge Andrésspa
dc.contributor.authorContreras Ordoñez, Victor Hugospa
dc.contributor.researchgroupMindLabspa
dc.date.accessioned2021-01-20T17:04:11Zspa
dc.date.available2021-01-20T17:04:11Zspa
dc.date.issued2019-10-31spa
dc.description.abstractLa búsqueda y recuperación de datos multimodales es una importante tarea dentro del campo de búsqueda y recuperación de información, donde las consultas y los elementos de la base de datos objetivo están representados por un conjunto de modalidades, donde cada una de ellas captura un aspecto de un fenómeno de interés. Cada modalidad contiene información complementaria y común a otras modalidades. Con el fin de tomar ventaja de la información adicional distribuida a través de las distintas modalidades han sido desarrollados muchos algoritmos y métodos que utilizan las propiedades estadísticas en los datos multimodales para encontrar correlaciones implícitas, otros aprenden a calcular distancias heterogéneas, otros métodos aprenden a proyectar los datos desde el espacio de entrada hasta un espacio semántico común, donde las diferentes modalidades son comparables y se puede construir un ranking a partir de ellas. En esta tesis se presenta el diseño de un sistema para la búsqueda y recuperación de información multimodal que aprende varias proyecciones no lineales a espacios semánticos latentes donde las distintas modalidades son representadas en conjunto y es posible realizar comparaciones y medidas de similitud para construir rankings multimodales. Adicionalmente se propone un método kernelizado para la proyección de datos a un espacio semántico latente usando la información de las etiquetas como método de supervisión para construir índice multimodal que integra los datos multimodales y la información de las etiquetas; este método puede proyectar los datos a tres diferentes espacios semánticos donde varias configuraciones de búsqueda y recuperación de información pueden ser aplicadas. El sistema y el método propuestos fueron evaluados en un conjunto de datos compuesto por casos médicos, donde cada caso consta de una imagen de tejido prostático, un reporte de texto del patólogo y un valor de Gleason score como etiqueta de supervisión. Combinando la información multimodal y la información en las etiquetas se generó un índice multimodal que se utilizó para realizar la tarea de búsqueda y recuperación de información por contenido obteniendo resultados sobresalientes. Las proyecciones no-lineales permiten al modelo una mayor flexibilidad y capacidad de representación. Sin embargo calcular estas proyecciones no-lineales en un conjunto de datos enorme es computacionalmente costoso, para reducir este costo y habilitar el modelo para procesar datos a gran escala, la técnica del budget fue utilizada, mostrando un buen compromiso entre efectividad y velocidad.spa
dc.description.abstractMultimodal information retrieval is an information retrieval sub-task where queries and database target elements are composed of several modalities or views. A modality is a representation of complex phenomena, captured and measured by different sensors or information sources, each one encodes some information about it. Each modality representation contains complementary and shared information about the phenomenon of interest, this additional information can be used to improve the information retrieval process. Several methods have been developed to take advantage of additional information distributed across different modalities. Some of them exploit statistical properties in multimodal data to find correlations and implicit relationships, others learn heterogeneous distance functions, and others learn linear and non-linear projections that transform data from the original input space to a common latent semantic space where different modalities are comparable. In spite of the attention dedicated to this issue, multimodal information retrieval is still an open problem. This thesis presents a multimodal information retrieval system designed to learn several mapping functions to transform multimodal data to a latent semantic space, where different modalities are combined and can be compared to build a multimodal ranking and perform a multimodal information retrieval task. Additionally, a multimodal kernelized latent semantic embedding method is proposed to construct a supervised multimodal index, integrating multimodal data and label supervision. This method can perform mappings to three different spaces where some information retrieval task setups can be performed. The proposed system and method were evaluated in a multimodal medical case-based retrieval task where data is composed of whole-slide images of prostate tissue samples, pathologist’s text report and Gleason score as a supervised label. Multimodal data and labels were combined to produce a multimodal index. This index was used to retrieve multimodal information and achieves outstanding results compared with previous works on this topic. Non-linear mappings provide more flexibility and representation capacity to the proposed model. However, constructing the non-linear mapping in a large dataset using kernel methods can be computationally costly. To reduce the cost and allow large scale applications, the budget technique was introduced, showing good performance between speed and effectiveness.spa
dc.description.additionalLínea de investigación: Ciencias de la computaciónspa
dc.description.degreelevelMaestríaspa
dc.description.projectJóvenes investigadores 761/2016spa
dc.description.sponsorshipCOLCIENCIASspa
dc.format.extent66spa
dc.format.mimetypeapplication/pdfspa
dc.identifier.citationContreras, V. H. (2019) Multimodal Non-linear Latent Semantic Method for Information Retrieval [Master's thesis, Universidad Nacional de Colombia]. SINABspa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/78848
dc.language.isoengspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Bogotáspa
dc.publisher.programBogotá - Ingeniería - Maestría en Ingeniería - Ingeniería de Sistemas y Computaciónspa
dc.relation.referencesAbadi, Mart´ın: TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. 2015. – Software available from tensorflow.orgspa
dc.relation.referencesArevalo, John ; Solorio, Thamar ; Montes-y Gomez ´ , Manuel ; Gonzalez ´ , Fabio A.: Gated multimodal units for information fusion. In: arXiv preprint arXiv:1702.01992 (2017)spa
dc.relation.referencesArora, Sanjeev ; Ge, Rong ; Kannan, Ravi ; Moitra, Ankur: Computing a nonnegative matrix factorization—Provably. In: SIAM Journal on Computing 45 (2016), Nr. 4, S. 1582–1611spa
dc.relation.referencesIn: Beitzel, Steven M. ; Jensen, Eric C. ; Frieder, Ophir: GMAP. Boston, MA : Springer US, 2009, S. 1256. – ISBN 978–0–387–39940–9spa
dc.relation.referencesIn: Benesty, Jacob ; Chen, Jingdong ; Huang, Yiteng ; Cohen, Israel: Pearson Correlation Coefficient. Berlin, Heidelberg : Springer Berlin Heidelberg, 2009, S. 1–4. – ISBN 978–3–642–00296–0spa
dc.relation.referencesBottou, L´eon: Large-scale machine learning with stochastic gradient descent. In: Proceedings of COMPSTAT’2010. Springer, 2010, S. 177–186spa
dc.relation.referencesBottou, L´eon ; Cun, Yann L.: Large scale online learning. In: Advances in neural information processing systems, 2004, S. 217–224spa
dc.relation.referencesBottou, L´eon ; Murata, Noboru: Stochastic approximations and efficient learning. In: The Handbook of Brain Theory and Neural Networks, Second edition,. The MIT Press, Cambridge, MA (2002)spa
dc.relation.referencesBottou, L´eon ; Le Cun, Yann: On-line learning for very large data sets. In: Applied stochastic models in business and industry 21 (2005), Nr. 2, S. 137–151spa
dc.relation.referencesBozzon, Alessandro ; Fraternali, Piero: Multimedia and multimodal information retrieval. In: Search Computing. Springer, 2010, S. 135–155spa
dc.relation.references] Bruno, Eric ; Marchand-Maillet, Stephane: Multiview clustering: a late fusion approach using latent models. In: Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval, 2009, S. 736–737spa
dc.relation.referencesCaicedo, Juan C. ; Gonzalez ´ , Fabio A.: Online Matrix Factorization for Multimodal Image Retrieval. (2012), S. 340–347spa
dc.relation.referencesCavallanti, Giovanni ; Cesa-Bianchi, Nicolo ; Gentile, Claudio: Tracking the best hyperplane with a simple budget perceptron. In: Machine Learning 69 (2007), Nr. 2-3, S. 143–167spa
dc.relation.referencesChitta, Radha ; Jin, Rong ; Jain, Anil K.: Efficient kernel clustering using random fourier features. In: 2012 IEEE 12th International Conference on Data Mining IEEE, 2012, S. 161–170spa
dc.relation.referencesContreras, Victor H. ; Lara, Juan S. ; Perdomo, Oscar J. ; Gonzalez ´ , Fabio A.: Supervised online matrix factorization for histopathological multimodal retrieval. In: 14th International Symposium on Medical Information Processing and Analysis Bd. 10975 International Society for Optics and Photonics, 2018, S. 109750Yspa
dc.relation.referencesCosta Pereira, Jose ; Coviello, Emanuele: On the role of correlation and abstraction in cross-modal multimedia retrieval. In: IEEE Transactions on Pattern Analysis and Machine Intelligence 36 (2014), Nr. 3, S. 521–535. – ISBN 0162–8828 VO – 36spa
dc.relation.referencesDe Lathauwer, Lieven ; De Moor, Bart ; Vandewalle, Joos: A multilinear singular value decomposition. In: SIAM journal on Matrix Analysis and Applications 21 (2000), Nr. 4, S. 1253–1278spa
dc.relation.referencesDeerwester, Scott ; Dumais, Susan T. ; Furnas, George W. ; Landauer, Thomas K. ; Harshman, Richard: Indexing by latent semantic analysis. In: Journal of the American society for information science 41 (1990), Nr. 6, S. 391–407spa
dc.relation.referencesDekel, Ofer ; Shalev-Shwartz, Shai ; Singer, Yoram: The Forgetron: A kernelbased perceptron on a fixed budget. In: Advances in neural information processing systems, 2006, S. 259–266spa
dc.relation.referencesDepeursinge, Adrien ; Muller ¨ , Henning: Fusion techniques for combining textual and visual information retrieval. In: ImageCLEF. Springer, 2010, S. 95–114spa
dc.relation.referencesEbert, Sandra ; Fritz, Mario ; Schiele, Bernt: Semi-supervised learning on a budget: scaling up to large datasets. In: Asian Conference on Computer Vision Springer, 2012, S. 232–245spa
dc.relation.referencesFeng, Fangxiang ; Li, Ruifan ; Wang, Xiaojie: Deep correspondence restricted Boltzmann machine for cross-modal retrieval. In: Neurocomputing 154 (2015), S. 50–60spa
dc.relation.referencesFeng, Fangxiang ; Wang, Xiaojie ; Li, Ruifan: Cross-modal retrieval with correspondence autoencoder. In: Proceedings of the 22nd ACM international conference on Multimedia, 2014, S. 7–16spa
dc.relation.referencesFowlkes, Charless ; Belongie, Serge ; Chung, Fan ; Malik, Jitendra: Spectral grouping using the Nystrom method. In: IEEE transactions on pattern analysis and machine intelligence 26 (2004), Nr. 2, S. 214–225spa
dc.relation.referencesFowlkes, Charless ; Belongie, Serge ; Malik, Jitendra: Efficient spatiotemporal grouping using the nystrom method. In: Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001 Bd. 1 IEEE, 2001, S. I–Ispa
dc.relation.referencesGhosh, Payel ; Antani, Sameer ; Long, L R. ; Thoma, George R.: Review of medical image retrieval systems and future directions. In: 2011 24th International Symposium on Computer-Based Medical Systems (CBMS) IEEE, 2011, S. 1–6spa
dc.relation.referencesGillis, Nicolas: Introduction to nonnegative matrix factorization. In: arXiv preprint arXiv:1703.00663 (2017)spa
dc.relation.referencesGonzalez ´ , Fabio A. ; Caicedo, Juan C. ; Nasraoui, Olfa ; Ben-Abdallah, Jaafar: NMF-based multimodal image indexing for querying by visual example. In: CIVR ACM, 2010, S. 366–373spa
dc.relation.referencesGower, J C.: Properties of Euclidean and non-Euclidean distance matrices. In: Linear Algebra and its Applications 67 (1985), S. 81–97. – ISSN 0024–3795spa
dc.relation.referencesGulli, Antonio ; Pal, Sujit: Deep Learning with Keras. Packt Publishing Ltd, 2017spa
dc.relation.referencesGunes, Hatice ; Piccardi, Massimo: Affect recognition from face and body: early fusion vs. late fusion. In: 2005 IEEE international conference on systems, man and cybernetics Bd. 4 IEEE, 2005, S. 3437–3443spa
dc.relation.referencesGupta, Amarnath ; Jain, Ramesh: Visual information retrieval. In: Communications of the ACM 40 (1997), Nr. 5, S. 70–78spa
dc.relation.referencesHan, Jiawei ; Kamber, Micheline ; Pei, Jian: 2 - Getting to Know Your Data. In: Han, Jiawei (Hrsg.) ; Kamber, Micheline (Hrsg.) ; Pei, Jian (Hrsg.): Data Mining (Third Edition). Third Edit. Boston : Morgan Kaufmann, 2012 (The Morgan Kaufmann Series in Data Management Systems). – ISBN 978–0–12–381479–1, S. 39–82spa
dc.relation.referencesHe, Jianfeng ; Ma, Bingpeng ; Wang, Shuhui ; Liu, Yugui ; Huang, Qingming: Cross-modal Retrieval by Real Label Partial Least Squares. In: Proceedings of the 2016 ACM on Multimedia Conference ACM, 2016, S. 227–231spa
dc.relation.referencesHernando, Antonio ; Bobadilla, Jes´us ; Ortega, Fernando: A non negative matrix factorization for collaborative filtering recommender systems based on a Bayesian probabilistic model. In: Knowledge-Based Systems 97 (2016), S. 188–202spa
dc.relation.referencesHofmann, Thomas: Probabilistic latent semantic analysis. In: Proceedings of the Fifteenth conference on Uncertainty in artificial intelligence Morgan Kaufmann Publishers Inc., 1999, S. 289–296spa
dc.relation.referencesHofmann, Thomas: Unsupervised learning by probabilistic latent semantic analysis. In: Machine learning 42 (2001), Nr. 1-2, S. 177–196spa
dc.relation.referencesHoyer, Patrik O.: Non-negative matrix factorization with sparseness constraints. In: Journal of machine learning research 5 (2004), Nr. Nov, S. 1457–1469spa
dc.relation.referencesJohnson, Rie ; Zhang, Tong: Accelerating stochastic gradient descent using predictive variance reduction. In: Advances in neural information processing systems, 2013, S. 315–323spa
dc.relation.referencesKludas, Jana ; Bruno, Eric ; Marchand-Maillet, Stephane: Information fusion in multimedia information retrieval. In: International Workshop on Adaptive Multimedia Retrieval Springer, 2007, S. 147–159spa
dc.relation.referencesKolda, Tamara G. ; O’leary, Dianne P.: A semidiscrete matrix decomposition for latent semantic indexing information retrieval. In: ACM Transactions on Information Systems (TOIS) 16 (1998), Nr. 4, S. 322–346spa
dc.relation.referencesKorenius, Tuomo ; Laurikkala, Jorma ; Juhola, Martti: On principal component analysis, cosine and Euclidean measures in information retrieval. In: Information Sciences 177 (2007), Nr. 22, S. 4893–4905spa
dc.relation.referencesKumar, Sanjiv ; Mohri, Mehryar ; Talwalkar, Ameet: Ensemble nystrom method. In: Advances in Neural Information Processing Systems, 2009, S. 1060–1068spa
dc.relation.referencesLahat, Dana ; Adali, T¨ulay ; Jutten, Christian: Multimodal data fusion: an overview of methods, challenges, and prospects. In: Proceedings of the IEEE 103 (2015), Nr. 9, S. 1449–1477spa
dc.relation.referencesLandauer, Thomas K. ; Foltz, Peter W. ; Laham, Darrell: An introduction to latent semantic analysis. In: Discourse processes 25 (1998), Nr. 2-3, S. 259–284spa
dc.relation.referencesLau, Jey H. ; Baldwin, Timothy: An empirical evaluation of doc2vec with practical insights into document embedding generation. In: arXiv preprint arXiv:1607.05368 (2016)spa
dc.relation.referencesLee, Daniel D. ; Seung, H S.: Learning the parts of objects by non-negative matrix factorization. In: Nature 401 (1999), Nr. 6755, S. 788–791spa
dc.relation.referencesLee, Daniel D. ; Seung, H S.: Algorithms for non-negative matrix factorization. In: Advances in neural information processing systems, 2001, S. 556–562spa
dc.relation.referencesLi, Chao ; Deng, Cheng ; Li, Ning ; Liu, Wei ; Gao, Xinbo ; Tao, Dacheng: Selfsupervised adversarial hashing networks for cross-modal retrieval. In: Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, S. 4242–4251spa
dc.relation.referencesLi, Zhongyu ; Zhang, Xiaofan ; Muller ¨ , Henning ; Zhang, Shaoting: Large-scale retrieval for medical image analytics: A comprehensive review. In: Medical image analysis 43 (2018), S. 66–84spa
dc.relation.referencesLiu, Dong ; Lai, Kuan-Ting ; Ye, Guangnan ; Chen, Ming-Syan ; Chang, Shih-Fu: Sample-specific late fusion for visual category recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2013, S. 803–810spa
dc.relation.referencesLiu, Wei ; Wang, Jun ; Ji, Rongrong ; Jiang, Yu-Gang ; Chang, Shih-Fu: Supervised hashing with kernels. In: Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on IEEE, 2012, S. 2074–2081spa
dc.relation.referencesLiu, Wei ; Wang, Jun ; Kumar, Sanjiv ; Chang, Shih-Fu: Hashing with graphs. In: Proceedings of the 28th international conference on machine learning (ICML-11) Citeseer, 2011, S. 1–8spa
dc.relation.referencesLiu, Xinwang ; Zhu, Xinzhong ; Li, Miaomiao ; Wang, Lei ; Tang, Chang ; Yin, Jianping ; Shen, Dinggang ; Wang, Huaimin ; Gao, Wen: Late fusion incomplete multi-view clustering. In: IEEE transactions on pattern analysis and machine intelligence 41 (2018), Nr. 10, S. 2410–2423spa
dc.relation.referencesMa, Lei ; Li, Hongliang ; Meng, Fanman ; Wu, Qingbo ; Ngan, King N.: Global and local semantics-preserving based deep hashing for cross-modal retrieval. In: Neurocomputing 312 (2018), S. 49–62spa
dc.relation.referencesManning, Christopher ; Raghavan, Prabhakar ; Schutze ¨ , Hinrich: Introduction to information retrieval. In: Natural Language Engineering 16 (2010), Nr. 1, S. 100–103spa
dc.relation.referencesManning, Christopher D. ; Raghavan, Prabhakar ; Schutze ¨ , Hinrich: Ch. 1 - Boolean retrieval. In: Introduction to Information Retrieval (2009), Nr. c, S. 1–18. – ISBN 0521865719spa
dc.relation.referencesMorvant, Emilie ; Habrard, Amaury ; Ayache, St´ephane: Majority vote of diverse classifiers for late fusion. In: Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR) Springer, 2014, S. 153–162spa
dc.relation.referencesMourao˜ , Andr´e ; Martins, Fl´avio ; Magalhaes ˜ , Jo˜ao: Multimodal medical information retrieval with unsupervised rank fusion. In: Computerized Medical Imaging and Graphics 39 (2015), S. 35–45spa
dc.relation.referencesMuller ¨ , Henning ; Michoux, Nicolas ; Bandon, David ; Geissbuhler, Antoine: A review of content-based image retrieval systems in medical applications - Clinical benefits and future directions. In: International Journal of Medical Informatics 73 (2004), Nr. 1, S. 1–23. – ISBN 1386–5056spa
dc.relation.referencesMuller ¨ , Henning ; Michoux, Nicolas ; Bandon, David ; Geissbuhler, Antoine: A review of content-based image retrieval systems in medical applications-clinical benefits and future directions. In: International journal of medical informatics 73 (2004), Nr. 1, S. 1–23spa
dc.relation.referencesNiblack, Carlton W. ; Barber, Ron ; Equitz, Will ; Flickner, Myron D. ; Glasman, Eduardo H. ; Petkovic, Dragutin ; Yanker, Peter ; Faloutsos, Christos ; Taubin, Gabriel: QBIC project: querying images by content, using color, texture, and shape. In: Storage and retrieval for image and video databases Bd. 1908 International Society for Optics and Photonics, 1993, S. 173–187spa
dc.relation.referencesPeng, Xiaojiang ; Wang, Limin ; Wang, Xingxing ; Qiao, Yu: Bag of visual words and fusion methods for action recognition: Comprehensive study and good practice. In: Computer Vision and Image Understanding 150 (2016), S. 109–125spa
dc.relation.referencesPeng, Yuxin ; Huang, Xin ; Qi, Jinwei: Cross-media shared representation by hierarchical learning with multiple deep networks. In: IJCAI, 2016, S. 3846–3853spa
dc.relation.referencesPereira, Jose C. ; Coviello, Emanuele ; Doyle, Gabriel ; Rasiwasia, Nikhil ; Lanckriet, Gert R. ; Levy, Roger ; Vasconcelos, Nuno: On the role of correlation and abstraction in cross-modal multimedia retrieval. In: IEEE Transactions on Pattern Analysis and Machine Intelligence 36 (2014), Nr. 3, S. 521–535spa
dc.relation.referencesRahimi, Ali ; Recht, Benjamin: Random features for large-scale kernel machines. In: Advances in neural information processing systems, 2008, S. 1177–1184spa
dc.relation.referencesRasiwasia, Nikhil ; Costa Pereira, Jose ; Coviello, Emanuele ; Doyle, Gabriel: A new approach to cross-modal multimedia retrieval. In: Mm (2010), S. 251–260. ISBN 9781605589336spa
dc.relation.referencesRastegar, Sarah ; Soleymani, Mahdieh ; Rabiee, Hamid R. ; Mohsen Shojaee, Seyed: Mdl-cw: A multimodal deep learning framework with cross weights. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, S. 2601–2609spa
dc.relation.referencesScholkopf ¨ , Bernhard ; Burges, Christopher J. ; Smola, Alexander J. [u. a.]: Advances in kernel methods: support vector learning. MIT press, 1999spa
dc.relation.referencesShalev-Shwartz, Shai [u. a.]: Online learning and online convex optimization. In: Foundations and TrendsR in Machine Learning 4 (2012), Nr. 2, S. 107–194spa
dc.relation.referencesShao, Jie ; Wang, Leiquan ; Zhao, Zhicheng ; Cai, Anni [u. a.]: Deep canonical correlation analysis with progressive and hypergraph learning for cross-modal retrieval. In: Neurocomputing 214 (2016), S. 618–628spa
dc.relation.referencesShawe-Taylor, John ; Cristianini, Nello [u. a.]: Kernel methods for pattern analysis. Cambridge university press, 2004spa
dc.relation.referencesSinghal, Amit [u. a.]: Modern information retrieval: A brief overview. In: IEEE Data Eng. Bull. 24 (2001), Nr. 4, S. 35–43spa
dc.relation.referencesSnoek, Cees G. ; Worring, Marcel ; Smeulders, Arnold W.: Early versus late fusion in semantic video analysis. In: Proceedings of the 13th annual ACM international conference on Multimedia ACM, 2005, S. 399–402spa
dc.relation.referencesSrivastava, Nitish ; Salakhutdinov, Ruslan: Learning representations for multimodal data with deep belief nets. In: International conference on machine learning workshop Bd. 79, 2012spa
dc.relation.referencesSrivastava, Nitish ; Salakhutdinov, Russ R.: Multimodal learning with deep boltzmann machines. In: Advances in neural information processing systems, 2012, S. 2222–2230spa
dc.relation.referencesdel Toro, Oscar J. ; Atzori, Manfredo ; Otalora ´ , Sebastian ; Andersson, Mats ; Euren´ , Kristian ; Hedlund, Martin ; Ronnquist ¨ , Peter ; Muller ¨ , Henning: Convolutional neural networks for an automatic classification of prostate tissue slides with high-grade gleason score. In: Medical Imaging 2017: Digital Pathology Bd. 10140 International Society for Optics and Photonics, 2017, S. 101400Ospa
dc.relation.referencesJimenez-del Toro, Oscar ; Otalora ´ , Sebastian ; Andersson, Mats ; Euren´ , Kristian ; Hedlund, Martin ; Rousson, Mikael ; Muller ¨ , Henning ; Atzori, Manfredo: Analysis of histopathology images: From traditional machine learning to deep learning. In: Biomedical Texture Analysis. Elsevier, 2018, S. 281–314spa
dc.relation.referencesJimenez-del Toro, Oscar ; Otalora ´ , Sebastian ; Atzori, Manfredo ; Muller ¨ , Henning: Deep Multimodal Case–Based Retrieval forLarge Histopathology Datasets. In: Wu, Guorong (Hrsg.) ; Munsell, Brent C. (Hrsg.) ; Zhan, Yiqiang (Hrsg.) ; Bai, Wenjia (Hrsg.) ; Sanroma, Gerard (Hrsg.) ; Coupe´, Pierrick (Hrsg.): Patch-Based Techniques in Medical Imaging. Cham : Springer International Publishing, 2017. – ISBN 978–3–319–67434–6, S. 149–157spa
dc.relation.referencesTREC: trec eval Evaluation Report. – Forschungsberichtspa
dc.relation.referencesVanegas, Jorge A.: Large-scale non-linear multimodal semantic embedding. Junio 2018. – Doctor en Ingenier´ıa. L´ınea de investigaci´on: Ciencias de la computaci´onspa
dc.relation.referencesVanegas, Jorge A. ; Escalante, Hugo J. ; Gonzalez ´ , Fabio A.: Semi-supervised Online Kernel Semantic Embedding for Multi-label Annotation. In: Mendoza, Marcelo (Hrsg.) ; Velast´ın, Sergio (Hrsg.): Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. Cham : Springer International Publishing, 2018. – ISBN 978–3–319–75193–1, S. 693–701spa
dc.relation.referencesVavasis, Stephen A.: On the complexity of nonnegative matrix factorization. In: SIAM Journal on Optimization 20 (2010), Nr. 3, S. 1364–1377spa
dc.relation.referencesWang, Daixin ; Cui, Peng ; Ou, Mingdong ; Zhu, Wenwu: Learning compact hash codes for multimodal representations using orthogonal deep structure. In: IEEE Transactions on Multimedia 17 (2015), Nr. 9, S. 1404–1416spa
dc.relation.referencesWang, Zhuang ; Vucetic, Slobodan: Twin vector machines for online learning on a budget. In: Proceedings of the 2009 SIAM International Conference on Data Mining SIAM, 2009, S. 906–917spa
dc.relation.referencesWu, Lin ; Wang, Yang ; Shao, Ling: Cycle-consistent deep generative hashing for cross-modal retrieval. In: IEEE Transactions on Image Processing 28 (2018), Nr. 4, S. 1602–1612spa
dc.relation.referencesXu, Wei ; Liu, Xin ; Gong, Yihong: Document clustering based on non-negative matrix factorization. In: Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval ACM, 2003, S. 267–273spa
dc.relation.referencesYang, Tianbao ; Li, Yu-Feng ; Mahdavi, Mehrdad ; Jin, Rong ; Zhou, Zhi-Hua: Nystr¨om method vs random fourier features: A theoretical and empirical comparison. In: Advances in neural information processing systems, 2012, S. 476–484spa
dc.relation.referencesYe, Guangnan ; Liu, Dong ; Jhuo, I-Hong ; Chang, Shih-Fu: Robust late fusion with rank minimization. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition IEEE, 2012, S. 3021–3028spa
dc.relation.referencesIn: Zhang, Ethan ; Zhang, Yi: Average Precision. Boston, MA : Springer US, 2009, S. 192–193. – ISBN 978–0–387–39940–9spa
dc.relation.referencesZhang, Jian ; Peng, Yuxin ; Yuan, Mingkuan: Unsupervised generative adversarial cross-modal hashing. In: Thirty-Second AAAI Conference on Artificial Intelligence, 2018spa
dc.relation.referencesZhang, Xiaofan ; Dou, Hang ; Ju, Tao ; Xu, Jun ; Zhang, Shaoting: Fusing heterogeneous features from stacked sparse autoencoder for histopathological image analysis. In: IEEE journal of biomedical and health informatics 20 (2016), Nr. 5, S. 1377–1383spa
dc.relation.referencesZhang, Zhong ; Qin, Zhili ; Li, Peiyan ; Yang, Qinli ; Shao, Junming: Multi-view Discriminative Learning via Joint Non-negative Matrix Factorization. In: International Conference on Database Systems for Advanced Applications Springer, 2018, S. 542–557spa
dc.relation.referencesZheng, Liang ; Wang, Shengjin ; Tian, Lu ; He, Fei ; Liu, Ziqiong ; Tian, Qi: Queryadaptive late fusion for image search and person re-identification. In: Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, S. 1741–1750spa
dc.relation.referencesZong, Linlin ; Zhang, Xianchao ; Zhao, Long ; Yu, Hong ; Zhao, Qianli: Multi-view clustering via multi-manifold regularized non-negative matrix factorization. In: Neural Networks 88 (2017), S. 74–89spa
dc.rightsDerechos reservados - Universidad Nacional de Colombiaspa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseAtribución-SinDerivadas 4.0 Internacionalspa
dc.rights.spaAcceso abiertospa
dc.rights.urihttp://creativecommons.org/licenses/by-nd/4.0/spa
dc.subject.ddc000 - Ciencias de la computación, información y obras generales::004 - Procesamiento de datos Ciencia de los computadoresspa
dc.subject.proposalRecuperación de información multimodalspa
dc.subject.proposalMultimodal information retrievaleng
dc.subject.proposalMultimodal latent semantic embeddingeng
dc.subject.proposalIncrustación semántica latente multimodalspa
dc.subject.proposalMatrix factorizationeng
dc.subject.proposalFactorización matricialspa
dc.subject.proposalMétodos del kernelspa
dc.subject.proposalKernel methodseng
dc.subject.proposalMultimodal fusioneng
dc.subject.proposalFusión multimodalspa
dc.subject.proposalMultimodal information retrieval systemeng
dc.subject.proposalSistema de recuperación de información multimodalspa
dc.titleMultimodal non-linear latent semantic method for information retrievalspa
dc.typeTrabajo de grado - Maestríaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/masterThesisspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
Thesis__Victor_UNAL_final.pdf
Tamaño:
1.82 MB
Formato:
Adobe Portable Document Format