31,136 Hits in 4.4 sec

Labelling Sulcal Graphs Across Indiviuals Using Multigraph Matching

N. Buskulic, F.X. Dupe, S. Takerkart, G. Auzias
2021 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI)  
Here, we quantitatively evaluated different graph matching approaches in both the pairwise and multigraph matching frameworks, on synthetic graphs simulating the structure and attributes distributions  ...  The problem of inter-individual comparison is of major importance in neuroimaging to detect patterns indicative of neurological pathology.  ...  The other pairwise algorithms IPF and SMAC provide poor results due to the presence of noise and outliers which they are not designed to deal with in such quantities.  ... 
doi:10.1109/isbi48211.2021.9434035 fatcat:rucsk7xv6bb6xhq2lkemh2scsm

LR-BCA: Label Ranking for Bridge Condition Assessment

Kai Wang, Tong Ruan, Faxiang Xie
2020 IEEE Access  
The HDC method firstly identifies all the label conflict examples, then iteratively filters out the noise.  ...  The use of machine learning for BCA can effectively solve the above problems. However, the large amount of label noise in the dataset severely affected the performance of the BCA model.  ...  [16] used random forest algorithm to identify noise. The agreement levels of decision trees in the forest were used for noise detection.  ... 
doi:10.1109/access.2020.3048419 fatcat:fwuvehjyynbenntq3fauarw43q

Semisupervised Association Learning Based on Partial Differential Equations for Sparse Representation of Image Class Attributes

Wei Song, Guang Hu, Liuqing OuYang, Zhenjie Zhu, Miaochao Chen
2021 Advances in Mathematical Physics  
Using the variational expectation-maximization algorithm, the whole generative process model can be inferred.  ...  In the framework of the image-like genus attribute model, data from different data sources are generated by their shared hidden space representation.  ...  Acknowledgments The study was supported by the National Social Science Fund (Grant No. 19BTY046) and Educational Commission of Hubei Province (Grant No. B2019197).  ... 
doi:10.1155/2021/4784411 fatcat:qhxwsag4rzfuvcazfzscrl2dbe

Instance Selection by Border Sampling in Multi-class Domains [chapter]

Guichong Li, Nathalie Japkowicz, Trevor J. Stocki, R. Kurt Ungar
2009 Lecture Notes in Computer Science  
Training sets contain much redundancy and noise in practical applications.  ...  The main problem is that previous approaches still suffer from the difficulty to produce effective samples for training classifiers.  ...  If the learning curve descends at Step 4, the algorithmic convergence is detected, and the previously learned result oD is returned as D′ at Step 5.  ... 
doi:10.1007/978-3-642-03348-3_22 fatcat:rr6kir7gkbgb5j7vj3ki423efm

From proteomic data to networks: statistics and methods reveal ciliary protein interaction landscape

Q Lu, K Koutroumpas, K Boldt, J Reeuwijk, N Katsanis, F Képès, R Roepman, M Ueffing, RB Russell
2015 Cilia  
After filtering hits from noise the constructed PIN is mined for protein clusters using a novel graph-clustering algorithm.  ...  Methods: The proposed framework consists of three steps. Initially, a revised Socio-Affinity algorithm [1] is applied to quantify the pairwise protein interaction affinities.  ...  After filtering hits from noise the constructed PIN is mined for protein clusters using a novel graph-clustering algorithm.  ... 
doi:10.1186/2046-2530-4-s1-p90 pmcid:PMC4519134 fatcat:mbftwqafkzh5vj42ukyoejtuh4

Relating ensemble diversity and performance: A study in class noise detection

Borut Sluban, Nada Lavrač
2015 Neurocomputing  
This paper explores the relation between ensemble diversity and noise detection performance in the context of ensemble-based class noise detection by studying different diversity measures on a range of  ...  It is shown that increased diversity of ensembles using the majority voting scheme does not lead to better noise detection performance and may even degrade the performance of heterogeneous noise detection  ...  Pairwise diversity measures Pairwise diversity measures are computed for each pair of classifiers or, in our case, noise detection algorithms from the set of L predictors that are used.  ... 
doi:10.1016/j.neucom.2014.10.086 fatcat:upb3eafjx5ax5b4srawecc2su4

On the Suitability of Fuzzy Rule-Based Classification Systems with Noisy Data

J. Saez, J. Luengo, F. Herrera
2012 IEEE transactions on fuzzy systems  
This paper analyzes the behavior of such systems with respect to classic crisp systems in the presence of noise.  ...  We study the performance of these systems and their robustness in terms of the performance degradation and the size of the classifiers when the noise level increases in training data.  ...  Sáez holds an FPU scholarship from the Spanish Ministry of Education and Science. J. Luengo holds a Post-Doctoral Research Fellowship at the University of Granada.  ... 
doi:10.1109/tfuzz.2012.2182774 fatcat:bjfnzgoxmrfzlekonakq5jnag4

Outlier Detection and Removal Algorithm in K-Means and Hierarchical Clustering

Anwesha Barai (Deb), Lopamudra Dey
2017 World Journal of Computer Application and Technology  
The goal of the project is to detect the outlier and remove the outliers to make the clustering more reliable.  ...  First apply clustering algorithm K-Means and Hierarchical clustering on a data set then find outliers from the each resulting clustering.  ...  After noise detection 36 data are detected as noise and Silhouette is increased. After applying particular algorithm all results are listed in above table.  ... 
doi:10.13189/wjcat.2017.050202 fatcat:6qy3pop52fatdkjdnuu5rzh4fy

Learning from Noisy Side Information by Generalized Maximum Entropy Model

Tianbao Yang, Rong Jin, Anil K. Jain
2010 International Conference on Machine Learning  
Although many algorithms have been developed to learn from side information, most of them assume perfect pairwise constraints.  ...  We consider the problem of learning from noisy side information in the form of pairwise constraints.  ...  Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of NSF, ONR and ARO.  ... 
dblp:conf/icml/YangJJ10 fatcat:4o7othdevjg4vmh77jsibdw2vm

Entity Embedding-based Anomaly Detection for Heterogeneous Categorical Events [article]

Ting Chen, Lu-An Tang, Yizhou Sun, Zhengzhang Chen, Kai Zhang
2016 arXiv   pre-print
Then we utilize the weighted pairwise interactions of different entity types to define the event probability.  ...  Using Noise-Contrastive Estimation with "context-dependent" noise distribution, our model can be learned efficiently regardless of the large event space.  ...  For each test event, it computes the conditional scores for all pairs of dependent and mutually exclusive subsets having up to k attributes, and combine the scores with a heuristic algorithm.  ... 
arXiv:1608.07502v1 fatcat:sawpikjxbnfkvifaclhxq5s3ji

Joint alignment of multispectral images via semidefinite programming

Yuanjie Zheng, Yu Wang, Wanzhen Jiao, Sujuan Hou, Yanju Ren, Maoling Qin, Dewen Hou, Chao Luo, Hong Wang, James Gee, Bojun Zhao
2017 Biomedical Optics Express  
This unique strategy takes a complete consideration of the information aggregated by all point-matching costs and enables the entire set of pairwise-image feature-mappings to be solved simultaneously and  ...  It solves a low-rank and semidefinite matrix that stores all pairwise-image feature-mappings by minimizing the total amount of point-to-point matching cost via a convex optimization of a semidefinite programming  ...  detected.  ... 
doi:10.1364/boe.8.000890 pmid:28270991 pmcid:PMC5330559 fatcat:licas7gojnbyvn4arssvx7iva4

Tackling the problem of classification with noisy data using Multiple Classifier Systems: Analysis of the performance and robustness

José A. Sáez, Mikel Galar, Julián Luengo, Francisco Herrera
2013 Information Sciences  
Traditional classifier learning algorithms build a unique classifier from the training data.  ...  of noise present in the dataset, but also on the way of creating diversity to build the final system.  ...  Acknowledgments Supported by the Spanish Ministry of Science and Technology under Projects TIN2011-28488 and TIN2010-15055, and also by the Regional Project P10-TIC-6858. José A.  ... 
doi:10.1016/j.ins.2013.06.002 fatcat:sjrf6vbbgra6hnx6mprr6ka76y

Realistic Smile Expression Recognition Approach Using Ensemble Classifier with Enhanced Bagging

Oday A. Hassen, Nur Azman Abu, Zaheera Zainal Abidin, Saad M. Darwish
2022 Computers Materials & Continua  
Still, the main causes of error in learning are due to noise, bias, and variance. Ensemble helps to minimize these factors.  ...  Our model takes advantage of the intrinsic association between face detection, smile, and other face features to alleviate the over-fitting issue on the limited training set and increase classification  ...  The average error rates (averaged over all datasets) derived from pairwise comparisons of the implemented algorithms as shown in Tab. 3.  ... 
doi:10.32604/cmc.2022.019125 fatcat:cp5ntlwgsrgotfjovcruaavnda

Evaporation from weighing precipitation gauges: impacts on automated gauge measurements and quality assurance methods

R. D. Leeper, J. Kochendorfer
2015 Atmospheric Measurement Techniques  
In general, the pairwise method that utilized a longer time series to smooth out sensor noise was more sensitive to gauge evaporation (−4.6% bias with respect to control) than the weighted-average method  ...  However, the use of evaporation suppressants is not always feasible due to environmental hazards and the added cost of maintenance, transport, and disposal of the gauge additive.  ...  This work was supported by NOAA through the Cooperative Institute for Climate and Satellites -North Carolina under Cooperative Agreement NA09NES4400006 in addition to support for USCRN from the NOAA Climate  ... 
doi:10.5194/amt-8-2291-2015 fatcat:clffaa6jgndpzgdc4vl6xaqwve

Concepts Not Alone: Exploring Pairwise Relationships for Zero-Shot Video Activity Recognition

Chuang Gan, Ming Lin, Yi Yang, Gerard Melo, Alexander G. Hauptmann
To handle noise and non-linearities in the ranking scores of the selected concepts, we propose a novel pairwise order matrix approach for score aggregation.  ...  Extensive experiments on the large-scale TRECVID Multimedia Event Detection data show the superiority of our approach.  ...  To alleviate the problems of noise and non-linearity of different ranking lists, we first convert the raw ranking scores of selected concepts into pairwise order matrices, in which each entry characterizes  ... 
doi:10.1609/aaai.v30i1.10466 fatcat:yyh4jqoz45fylnowvkcgwjvyiy
« Previous Showing results 1 — 15 out of 31,136 results