Filters








147,222 Hits in 2.6 sec

Kernel Topic Models [article]

Philipp Hennig, David Stern, Ralf Herbrich, Thore Graepel
2011 arXiv   pre-print
This allows documents to be associated with elements of a Hilbert space, admitting kernel topic models (KTM), modelling temporal, spatial, hierarchical, social and other structure between documents.  ...  The KTM can also be interpreted as a type of Gaussian process latent variable model, or as a topic model conditional on document features, uncovering links between earlier work in these areas.  ...  Inference in the kernel topic model is cubic in the number of documents.  ... 
arXiv:1110.4713v1 fatcat:klwrebjkqvakxo6n24d7iccudi

Topic Model Kernel Classification With Probabilistically Reduced Features

Vu Nguyen, Dinh Phung, Svetha Venkatesh
2021 Journal of Data Science  
In this paper, we describe the Topic Model Kernel (TMK), a topicbased kernel for Support Vector Machine classification on data being processed by probabilistic topic models.  ...  Probabilistic topic models have become a standard in modern machine learning to deal with a wide range of applications.  ...  performance.  Inverser Multiquadric Kernel: Figure 1 : 1 Probabilistic Topic Models.  ... 
doi:10.6339/jds.201507_13(3).0006 fatcat:5efq7yocmjfa3i3koqicn5vubm

Using Kernel Density Classifier with Topic Model and Cost Sensitive Learning for Automatic Text Categorization

Dwi Sianto Mansjur, Ted S. Wada, Biing Hwang Juang
2009 2009 10th International Conference on Document Analysis and Recognition  
The experimental results confirm the effectiveness of the framework to utilize the features from the topic model for cost-sensitive categorization. 2009 10th International Conference on Document Analysis  ...  This paper proposes a novel framework for automatic text categorization problem based on the kernel density classifier.  ...  Topic Model In this paper, we use LSA to obtain a topic model.  ... 
doi:10.1109/icdar.2009.145 dblp:conf/icdar/MansjurWJ09 fatcat:cb4cnpszwnh6xb42yk2srd74y4

A Gaussian Kernel-based Spatiotemporal Fusion Model for Agricultural Remote Sensing Monitoring

Yonglin Shen, Guoling Shen, Han Zhai, Chao Yang, Kunlun Qi
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Index Terms-Spatiotemporal fusion; Gaussian kernel; time series; normalized difference vegetation index (NDVI) Ⅰ.  ...  The experimental results show that GKSFM outperformed the comparative models in different proportions of cropland/non-cropland and different crop phenology.  ...  In this study, a Gaussian kernel-based spatiotemporal fusion model (GKSFM) was proposed to fuse high-resolution NDVI (Landsat) and low-resolution NDVI (MODIS) during the crop growing season to produce  ... 
doi:10.1109/jstars.2021.3066055 fatcat:kooy6scex5dd7eeqex7lh5vyey

Development of Kernel-Driven Models With Fixed Hotspot Width Under a General Modeling Framework in the Thermal Infrared Domain

Xiangyang Liu, Bo-Hui Tang, Zhao-Liang Li, Guofei Shang
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
nine kernel-driven models with different coefficient requirements.  ...  Under a general kernel-driven modeling framework proposed by Cao et al., by using three fixed-width hotspot kernels, and considering whether to combine two existing base shape kernels, this article proposed  ...  ACKNOWLEDGMENT This authors would like to thank CESBIO for providing us the DART model 2 and Prof. B. Cao for providing instructions on the use of DART model.  ... 
doi:10.1109/jstars.2021.3110208 fatcat:r5u4a7przfhtpombchr4eu26gu

A KERNEL METHOD BASED ON TOPIC MODEL FOR VERY HIGH SPATIAL RESOLUTION (VHSR) REMOTE SENSING IMAGE CLASSIFICATION

Linmei Wu, Li Shen, Zhipeng Li
2016 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
The new kernel method is based on spectral-spatial information and structure information as well, which is acquired from topic model, Latent Dirichlet Allocation model.  ...  The result shows that the overall accuracy of the spectral- and structure-based kernel method is 80 %, which is higher than the spectral-based kernel method, as well as the spectral- and spatial-based  ...  TOPIC MODEL Topic model is developed initially in text analysis domain for category and annotation.  ... 
doi:10.5194/isprsarchives-xli-b7-399-2016 fatcat:w3sec23hangalff33giqhk6if4

A KERNEL METHOD BASED ON TOPIC MODEL FOR VERY HIGH SPATIAL RESOLUTION (VHSR) REMOTE SENSING IMAGE CLASSIFICATION

Linmei Wu, Li Shen, Zhipeng Li
2016 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
The new kernel method is based on spectral-spatial information and structure information as well, which is acquired from topic model, Latent Dirichlet Allocation model.  ...  The result shows that the overall accuracy of the spectral- and structure-based kernel method is 80 %, which is higher than the spectral-based kernel method, as well as the spectral- and spatial-based  ...  TOPIC MODEL Topic model is developed initially in text analysis domain for category and annotation.  ... 
doi:10.5194/isprs-archives-xli-b7-399-2016 fatcat:3zxlscbt7vd45pruah62bqzp6i

Improving kernel-driven BRDF model for capturing vegetation canopy reflectance with large leaf inclinations

Shengbiao Wu, Jianguang Wen, Qinhuo Liu, Dongqin You, Gaofei Yin, Xinwen Lin
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Index Terms-Bidirectional reflectance distribution function (BRDF) /bidirectional reflectance factor (BRF), kernel-driven model, leaf inclination distribution, vegetation, volumetric scattering kernel.  ...  Subsequently, we improved the RTLSR model into a four-parameter version (RTLSRV4p) with a new volumetric scattering kernel derived from the assumption of vertical leaf inclination.  ...  Féret for providing the PROSAIL model code. The authors would also like to thank Dr. L.Y. Liu for sharing the valuable in situ canopy reflectance and SIF datasets.  ... 
doi:10.1109/jstars.2020.2987424 fatcat:7noloigfn5f6hfzgyog2ig7tmi

Gaussian Process Topic Models [article]

Amrudin Agovic, Arindam Banerjee
2012 arXiv   pre-print
We introduce Gaussian Process Topic Models (GPTMs), a new family of topic models which can leverage a kernel among documents while extracting correlated topics.  ...  Since GPTMs work with both a topic covariance matrix and a document kernel matrix, learning GPTMs involves a novel component-solving a suitable Sylvester equation capturing both topic and document dependencies  ...  The final Figure 1 : Gaussian Process Topic Model embedding will be based on both the kernel as well as the structure of the documents as determined by the topic model.  ... 
arXiv:1203.3462v1 fatcat:56jsfvyd3bgi7nrunjoprt3fbu

Scalable Generalized Dynamic Topic Models [article]

Patrick Jähnichen, Florian Wenzel, Marius Kloft, Stephan Mandt
2018 arXiv   pre-print
Dynamic topic models (DTMs) model the evolution of prevalent themes in literature, online media, and other forms of text over time.  ...  These dynamical priors make inference much harder than in regular topic models, and also limit scalability. In this paper, we present several new results around DTMs.  ...  If this prior is a Gaussian process, this leads to the kernel topic model (Hennig et al., 2012) or Gaussian process topic model (Agovic and Banerjee, 2012) .  ... 
arXiv:1803.07868v1 fatcat:v2irkbcbajahhkyjfh75pergay

A Composite Kernel Approach for Dialog Topic Tracking with Structured Domain Knowledge from Wikipedia

Seokhwan Kim, Rafael E. Banchs, Haizhou Li
2014 Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)  
This paper proposes a composite kernel approach for dialog topic tracking to utilize various types of domain knowledge obtained from Wikipedia.  ...  The experimental results show that our composite kernel approach can significantly improve the performances of topic tracking in mixed-initiative human-human dialogs.  ...  While the history sequence kernel enhanced the coverage of the model to detect topic transitions, the domain context tree kernel contributed to produce more precise outputs.  ... 
doi:10.3115/v1/p14-2004 dblp:conf/acl/KimBL14 fatcat:xolq3pzpwjdvdprklzb6jrciqi

EARTH OBSERVATION IMAGE SEMANTICS: LATENT DIRICHLET ALLOCATION BASED INFORMATION DISCOVERY

Mohammadi Asiyabi Reza, Datcu Mihai
2021 Zenodo  
In the present study, Latent Dirichlet Allocation is employed for semantic discovery in RS images and a novel kernel-based Bag of Visual Words model is proposed for land cover mapping.  ...  In the Kernel-based BOVW stage, the pixel-wise mid-level representation of the RS image is produced using the proposed kernel-based BOVW model.  ...  However, further investigation is necessary in future studies to evaluate the performance of the kernel-based BOVW model as a pixel-wise alteration for patch-based BOVW model.  ... 
doi:10.5281/zenodo.6220982 fatcat:ad2dh7ybnjbanmynsmzm7pk3mu

Supporting systematic reviews using LDA-based document representations

Yuanhan Mo, Georgios Kontonatsios, Sophia Ananiadou
2015 Systematic Reviews  
We apply Latent Dirichlet allocation (LDA), an unsupervised topic modelling approach, to automatically identify topics in a collection of studies.  ...  Methods: We explore the use of topic modelling methods to derive a more informative representation of studies.  ...  Figures 3 and 4 illustrate the results of using RBF and POLY kernel functions, respectively, in training BOW, topic-based models and TE-topic-based model on the youth development corpus.  ... 
doi:10.1186/s13643-015-0117-0 pmid:26612232 pmcid:PMC4662004 fatcat:3npam4hxlrfatimk56rh4iklri

Combining Thesaurus Knowledge and Probabilistic Topic Models [chapter]

Natalia Loukachevitch, Michael Nokel, Kirill Ivanov
2017 Lecture Notes in Computer Science  
If a general thesaurus, such as WordNet, is used, the thesaurus-based improvement of topic models can be achieved with excluding hyponymy relations in combined topic models.  ...  In this paper we present the approach of introducing thesaurus knowledge into probabilistic topic models.  ...  This study is supported by Russian Scientific Foundation in part concerning the combined approach uniting thesaurus information and probabilistic topic models (project N16-18-02074).  ... 
doi:10.1007/978-3-319-73013-4_6 fatcat:pz3p7qgwkffczci7fbphww4j7u

TPRM: A Topic-based Personalized Ranking Model for Web Search [article]

Minghui Huang, Wei Peng, Dong Wang
2021 arXiv   pre-print
In this paper, we propose a topic-based personalized ranking model (TPRM) that integrates user topical profile with pretrained contextualized term representations to tailor the general document ranking  ...  Experiments on the real-world dataset demonstrate that TPRM outperforms state-of-the-art ad-hoc ranking models and personalized ranking models significantly.  ...  Topic Number The number of topics is an important hyper-parameter in topic models.  ... 
arXiv:2108.06014v1 fatcat:wlnv744frvfsrexfocivyxbjzm
« Previous Showing results 1 — 15 out of 147,222 results