52,490 Hits in 5.4 sec

Pre-processing Large Spatial Data Sets with Bayesian Methods [chapter]

Saara Hyvönen, Esa Junttila, Marko Salmenkivi
Lecture Notes in Computer Science  
We demonstrate that a simple Bayesian modeling approach can be used in pre-processing large spatial data sets with missing or uncertain data.  ...  Our experiments on real and synthetic data show that conducting the pre-processing phase before applying conventional data mining methods, such as PCA, clustering or NMF, improves the results significantly  ...  In this paper we demonstrate that it is feasible to pre-process large spatial binary data sets with Bayesian methods and -what is the main point -to obtain good results in subsequent analyses.  ... 
doi:10.1007/978-3-540-74976-9_51 fatcat:kpvhc27abfgrplmfjhtycdfxlm

Fast Visual Trajectory Analysis Using Spatial Bayesian Networks

Thomas Liebig, Christine Körner, Michael May
2009 2009 IEEE International Conference on Data Mining Workshops  
Also recent developments in the area of trajectory data warehouses cannot be applied because the spatial correlations are lost during trajectory aggregation.  ...  More precisely, we build a Bayesian Network model using the Scalable Sparse Bayesian Network Learning (SSBNL) algorithm [1], which we improve to represent also negative correlations.  ...  We thank our project partner elis 1 for data preparation and provision of the application data.  ... 
doi:10.1109/icdmw.2009.44 dblp:conf/icdm/LiebigKM09 fatcat:sw66vymdrnaxrgmnvgc6ft6f3q

Modified Bayesian algorithm implemented in compressive sensing applied to spatially sampled GPR measurement under high clutter conditions

Riafeni Karlina, Motoyuki Sato
2018 Nonlinear Theory and Its Applications IEICE  
Full frequency information was used in pre-processing to suppress the noise and clutter in the experiment data.  ...  In the first step, the spatial sampling was directly conducted in the data acquisition process. In the second step, the frequency sampling was conducted offline during the signal processing.  ...  After pre-processing, the frequency data was downsampled by choosing a different set of 40 frequency points, or about 29% of the full frequency data, for each spatial point.  ... 
doi:10.1587/nolta.9.121 fatcat:uc23tngowrhwvbgznoml72mo7i

Bayesian Markov Chain Random Field Cosimulation for Improving Land Cover Classification Accuracy

Weidong Li, Chuanrong Zhang, Michael R. Willig, Dipak K. Dey, Guiling Wang, Liangzhi You
2014 Mathematical Geosciences  
It was tested using a series of expert-interpreted data sets and an image data set pre-classified by the supervised maximum likelihood (SML) algorithm. Results show that with the density  ...  The pre-classification can be performed using any convenient conventional method.  ...  Because a MCRF model is a spatial Bayesian model at the neighborhood level with sequential Bayesian updating on different nearest data, the incorporation of an auxiliary data set in a Co-MCRF model is  ... 
doi:10.1007/s11004-014-9553-y fatcat:6kgdnqny4jep3jezrvbuc457cm

Scalable Sparse Bayesian Network Learning for Spatial Applications

Thomas Liebig, Christine Körner, Michael May
2008 2008 IEEE International Conference on Data Mining Workshops  
Furthermore, we apply our method to German cities, evaluate the accuracy and analyse the runtime behaviour for different parameter settings.  ...  We introduce and examine a Bayesian Network Learning algorithm enabling us to handle the complexity and performance requirements of the spatial context.  ...  It uses the sparseness within the data by processing frequent sets of random variables.  ... 
doi:10.1109/icdmw.2008.124 dblp:conf/icdm/LiebigKM08 fatcat:w3ocuprgprhfninttoyxfugqv4

Integrating Recursive Bayesian Estimation with Support Vector Machine to Map Probability of Flooding from Multispectral Landsat Data

Chandrama Sarker, Luis Mejias Alvarez, Alan Woodley
2016 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA)  
Data with multiple spectral bands and large area coverage generally resulted in out of memory and the program failed to run the classification process.  ...  Fig. 3 3 shows the pre and post Bayesian flood probability map for two test data -Data 4 and Data 6.  ... 
doi:10.1109/dicta.2016.7797054 dblp:conf/dicta/SarkerAW16 fatcat:fhh7dv2lsbgrlkcrhntg3i4cem

Optimal Bayesian fusion of large hyperspectral astronomical observations

M. Petremand, A. Jalobeanu, C. Collet
2012 Statistical Methodology  
In this paper, we propose a new data fusion method that aims at reconstructing a single data cube from this large CCD (raw) data set (up to 100 GB for a MUSE acquisition session) taking into account acquisition  ...  The new hyperspectral observations will represent huge amount of scientific data (up to 1.2 GB per each MUSE raw acquisition) whose analysis requires the development of dedicated processing methods.  ...  Objects reconstructed with interpolation methods still exhibit blur compared to the Bayesian fusion result, especially for high band numbers where spatial and spectral resolutions in the observation set  ... 
doi:10.1016/j.stamet.2011.04.007 fatcat:3om6bdun7ve4hnp4fpjob5vh3e

Machine-learning for cluster analysis of localization microscopy data [article]

David J Williamson, Garth L Burn, Juliette Griffie, Daniel M Davis, Dylan M Owen
2018 bioRxiv   pre-print
Many existing computational approaches are also limited in their ability to process large-scale data-sets or to deal effectively with inhomogeneities in clustering.  ...  We demonstrate the performance on simulated data and experimental data of the kinase Csk and the adaptor PAG in both naive and pre-stimulated primary human T cell synapses.  ...  For high-density data (3000 points per μm²) the Bayesian method was unable to finish processing any of the 3 × 3 μm region subsets.  ... 
doi:10.1101/505719 fatcat:vs4jvtbayfa3tkhqrqtzvxqgfu

On the use of ICA for hyperspectral image analysis

A. Villa, J. Chanussot, C. Jutten, J. A. Benediktsson, S. Moussaoui
2009 2009 IEEE International Geoscience and Remote Sensing Symposium  
We propose a hierarchical approximation for the use of ICA as a pre-processing step for a Bayesian Positive Source Separation method.  ...  In recent years ICA has been largely studied by researchers from the signal processing community.  ...  The proposed method ICDA, which uses ICA before a Bayesian classification, shows interesting results and gives the best results for the ROSIS and AVIRIS data set with reduced training set.  ... 
doi:10.1109/igarss.2009.5417363 dblp:conf/igarss/VillaCJBM09 fatcat:vexpdrpkjvh3xcm3szf4fpel2q

Area-Correlated Spectral Unmixing Based on Bayesian Nonnegative Matrix Factorization

Xiawei Chen, Jing Yu, Weidong Sun
2013 Open Journal of Applied Sciences  
To solve the problem of the spatial correlation for adjacent areas in traditional spectral unmixing methods, we propose an area-correlated spectral unmixing method based on Bayesian nonnegative matrix  ...  In the proposed method, the spatial correlation property between two adjacent areas is expressed by a priori probability density function, and the endmembers extracted from one of the adjacent areas are  ...  After pre-processing, Bayesian nonnegative matrix factorization with the spatial correlation constraint, as described later in Section 2.2, will be applied to the current hyperspectral data to estimate  ... 
doi:10.4236/ojapps.2013.31b009 fatcat:si5g3wjkyvhqbooqy7a3umrefi

Sentinel-2 Satellite Imagery for Urban Land Cover Classification by Optimized Random Forest Classifier

Tianxiang Zhang, Jinya Su, Zhiyong Xu, Yulin Luo, Jiangyun Li
2021 Applied Sciences  
The improved random forest with Bayesian parameter optimization is compared against the support vector machine (SVM) and random forest (RF) with default hyperparameters by discriminating five land cover  ...  Therefore, this work aims to propose a systematic method to automatically tune the hyperparameters by Bayesian parameter optimization for the random forest classifier.  ...  Data Availability Statement: No new data were created or analyzed in this study. Data sharing is not applicable to this article. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/app11020543 fatcat:zwae46nh55d25ph4vdks5bw6oe

Mapping pre-European settlement vegetation at fine resolutions using a hierarchical Bayesian model and GIS

Hong S. He, Daniel C. Dey, Xiuli Fan, Mevin B. Hooten, John M. Kabrick, Christopher K. Wikle, Zhaofei Fan
2006 Plant Ecology  
In this study, we applied a hierarchical Bayesian approach that combines species and environment relationships and explicit spatial dependence to map GLO data.  ...  Thus, geographic information system and statistical inference methods to map GLO data and reconstruct historical vegetation are needed.  ...  Ultimately, our interest lies with the predictions of Y j , where j can exist on some other set of spatial location than our original data.  ... 
doi:10.1007/s11258-006-9216-2 fatcat:i2noptu6ojcmfhkuhrumaf6y4y

Two Approaches to Imputation and Adjustment of Air Quality Data from a Composite Monitoring Network

Alessio Pollice, Giovanna Jona Lasinio
2021 Journal of Data Science  
Preliminary analysis involved addressing several data problems, mainly: (i) an imputation techniques were considered to cope with the large number of missing data, due to both different working periods  ...  and prediction of spatial linear mixed effects models.  ...  Here our main concern is on two methods for pre-processing data recorded from an air quality monitoring network characterized by missing data and heterogeneity.  ... 
doi:10.6339/jds.2009.07(1).589 fatcat:d7s7ucy4sbewvim7salxbdgffu

Bayesian Optimization for Neuroimaging Pre-processing in Brain Age Classification and Prediction

Jenessa Lancaster, Romy Lorenz, Rob Leech, James H. Cole
2018 Frontiers in Aging Neuroscience  
Bayesian Optimization for Brain-Age Analysis parameters. Our results suggest that different pre-processing parameters are selected when optimization is conducted in specific contexts.  ...  This potentially motivates use of optimization techniques at many different points during the experimental process, which may improve statistical sensitivity and reduce opportunities for experimenter-led  ...  RLo, RLe, and JC developed the methods. JL, RLe, and JC analyzed and interpreted the data. JL, RLo, RLe, and JC drafted and revised the manuscript.  ... 
doi:10.3389/fnagi.2018.00028 pmid:29483870 pmcid:PMC5816033 fatcat:2u3efcteebandj7wze4rrlopym

Markov chain random fields, spatial Bayesian networks, and optimal neighborhoods for simulation of categorical fields [article]

Weidong Li, Chuanrong Zhang
2018 arXiv   pre-print
The Markov chain random field (MCRF) model/theory provides a non-linear spatial Bayesian updating solution at the neighborhood nearest data level for simulating categorical spatial variables.  ...  In the MCRF solution, the spatial dependencies among nearest data and the central random variable is a probabilistic directed acyclic graph that conforms to a neighborhood-based Bayesian network on spatial  ...  The data dependencies also indicate the sequential Bayesian updating process.  ... 
arXiv:1807.06111v2 fatcat:fxnctf45fzcylbgwej4qyfafhi
« Previous Showing results 1 — 15 out of 52,490 results