Filters








5,141 Hits in 6.8 sec

A Bayesian Hierarchical Model for Comparing Average F1 Scores

Dell Zhang, Jun Wang, Xiaoxue Zhao, Xiaoling Wang
2015 2015 IEEE International Conference on Data Mining  
Sensitive to the choice of prior distribution in the alternative model.  ...  To compare classifiers on any type of data, e.g., images. Figure: The probabilistic graphical model for estimating the uncertainty of average F 1 scores. k =1 c jk = n j .  ...  Experiments Summary The main contribution of this paper is a Bayesian estimation approach to assessing the uncertainty of average F 1 scores in multi-class text classification.  ... 
doi:10.1109/icdm.2015.44 dblp:conf/icdm/ZhangWZW15 fatcat:mcclumlbvzhjtpnzm5jp5vql7q

A XGBoost risk model via feature selection and Bayesian hyper-parameter optimization [article]

Yan Wang, Xuelei Sherry Ni
2019 arXiv   pre-print
TPE optimization shows a superiority over RS since it results in a significantly higher accuracy and a marginally higher AUC, recall and F1 score.  ...  The performance of XGBoost is compared to the traditionally utilized logistic regression (LR) model in terms of classification accuracy, area under the curve (AUC), recall, and F1 score obtained from the  ...  Compared with the rest of the FS methods, although Cluster achieves the second highest F1 score in XGB_TPE, it results the lowest F1 score in LR.  ... 
arXiv:1901.08433v1 fatcat:4w4xbt4xvfb6jbxcibvddaewxm

A XGBOOST RISK MODEL VIA FEATURE SELECTION AND BAYESIAN HYPER-PARAMETER OPTIMIZATION

Yan Wang, Xuelei Sherry Ni
2019 International Journal of Database Management Systems  
TPE optimization shows a superiority over RS since it results in a significantly higher accuracy and a marginally higher AUC, recall and F1 score.  ...  The performance of XGBoost is compared to the traditionally utilized logistic regression (LR) model in terms of classification accuracy, area under the curve (AUC), recall, and F1 score obtained from the  ...  XGB_TPE, which bases on Bayesian hyper-parameter optimization approach, achieves a higher accuracy, recall, and F1 score than XGB_RS that bases on a random trial-and-error process.  ... 
doi:10.5121/ijdms.2019.11101 fatcat:dhgatigasff4fne6hjsssyaqxa

Dynamic categorization of clinical research eligibility criteria by hierarchical clustering

Zhihui Luo, Meliha Yetisgen-Yildiz, Chunhua Weng
2011 Journal of Biomedical Informatics  
The J48 classifier yielded the best F1-score and the Bayesian Network classifier achieved the best learning efficiency.  ...  , recall, and F1-score) between the UMLS-based feature representation and the "bag of words" feature representation among five common classifiers in Weka, including J48, Bayesian Network, Naïve Bayesian  ...  We also thank Yalini Senathirajah for her help with the Hierarchical Clustering Explorer software.  ... 
doi:10.1016/j.jbi.2011.06.001 pmid:21689783 pmcid:PMC3183114 fatcat:gtpyli2fa5gktgph6yoqcdlfni

Nonparametric Bayesian Word Sense Induction

Xuchen Yao, Benjamin Van Durme
2011 Workshop on Graph-based Methods for Natural Language Processing  
We propose the use of a nonparametric Bayesian model, the Hierarchical Dirichlet Process (HDP), for the task of Word Sense Induction.  ...  Results are shown through comparison against Latent Dirichlet Allocation (LDA), a parametric Bayesian model employed by Brody and Lapata (2009) for this task.  ...  ) , a nonparametric Bayesian model.  ... 
dblp:conf/textgraphs/YaoD11 fatcat:oexsh6pdjnfgppgk3bmcdpaija

Empowering differential networks using Bayesian analysis

Jarod Smith, Mohammad Arashi, Andriëtte Bekker, Marton Karsai
2022 PLoS ONE  
A Bayesian approach for estimating DNs, from the classical viewpoint, is introduced with a computationally efficient threshold selection for graphical model determination.  ...  Differential networks (DN) are important tools for modeling the changes in conditional dependencies between multiple samples.  ...  Acknowledgments We would like to sincerely thank both anonymous reviewers for their generous comments on the manuscript.  ... 
doi:10.1371/journal.pone.0261193 pmid:35077451 pmcid:PMC8789149 fatcat:oshuwiu3ojbddpmqlypkg3rhdq

Multi-label Classification for Tree and Directed Acyclic Graphs Hierarchies [chapter]

Mallinali Ramírez-Corona, L. Enrique Sucar, Eduardo F. Morales
2014 Lecture Notes in Computer Science  
For NMLNP we compared several pruning approaches varying the pruning direction, pruning time and pruning condition.  ...  Additionally, we propose a new evaluation metric for hierarchical classifiers, that avoids the bias of current measures which favor conservative approaches when using NMLNP.  ...  and compared with various hierarchical classification techniques.  ... 
doi:10.1007/978-3-319-11433-0_27 fatcat:xlixcx5655eq3lmelqjzcyxgdm

Empowering Differential Networks Using Bayesian Analysis [article]

Jarod Smith, Mohammad Arashi, Andriette Bekker
2021 arXiv   pre-print
A Bayesian approach for estimating DNs, from the classical viewpoint, is introduced with a computationally efficient threshold selection for graphical model determination.  ...  Differential networks (DN) are important tools for modeling the changes in conditional dependencies between multiple samples.  ...  Hierarchical representation [12] propose a hierarchical representation of the graphical lasso prior Eq (3), using the Bayesian lasso formulation in [26] .  ... 
arXiv:2105.13584v1 fatcat:dm7v67lgpvbrxkfndm3voxt5k4

Bayesian models for Large-scale Hierarchical Classification

Siddharth Gopal, Yiming Yang, Bing Bai, Alexandru Niculescu-Mizil
2012 Neural Information Processing Systems  
This paper proposes a set of Bayesian methods to model hierarchical dependencies among class labels using multivariate logistic regression.  ...  A challenging problem in hierarchical classification is to leverage the hierarchical relations among classes for improving classification performance.  ...  A major part of work was accomplished while the first author was interning at NEC Labs, Princeton.  ... 
dblp:conf/nips/GopalYBN12 fatcat:dhb2ovrk3vadjivyydjxsv5g4y

CloudRCA: A Root Cause Analysis Framework for Cloud Computing Platforms [article]

Yingying Zhang, Zhengxiong Guan, Huajie Qian, Leili Xu, Hengbo Liu, Qingsong Wen, Liang Sun, Junwei Jiang, Lunting Fan, Min Ke
2021 arXiv   pre-print
The engineered features are then utilized in a Knowledge-informed Hierarchical Bayesian Network (KHBN) model to infer root causes with high accuracy and efficiency.  ...  Ablation study and comprehensive experimental comparisons demonstrate that, compared to existing frameworks, CloudRCA 1) consistently outperforms existing approaches in f1-score across different cloud  ...  To be precise, we compute a score for each log in a cluster based on its average distance to other logs within the same cluster: Score(𝑖) = 1 𝑛 − 1 𝑛 ∑︁ 𝑗=1 (1 − similarity(log 𝑖 , log 𝑗 )) where  ... 
arXiv:2111.03753v1 fatcat:yuo63z3xgzbi5kurvysnuoevcu

Partially Observed Maximum Entropy Discrimination Markov Networks

Jun Zhu, Eric P. Xing, Bo Zhang
2008 Neural Information Processing Systems  
In this paper, we present a partially observed Maximum Entropy Discrimination Markov Network (PoMEN) model that attempts to combine the advantages of Bayesian and margin based paradigms for learning Markov  ...  We develop an EM-style algorithm utilizing existing convex optimization algorithms for M 3 N as a subroutine.  ...  Due to space limitation, we only report average F1 and block instance accuracy. For ST1, PoMEN achieves better scores and lower variances than PoHCRF in both average F1 and block instance accuracy.  ... 
dblp:conf/nips/ZhuXZ08 fatcat:v3hmnoctffax3otjj3zvkssi3y

Personalizing Gesture Recognition Using Hierarchical Bayesian Neural Networks

Ajjen Joshi, Soumya Ghosh, Margrit Betke, Stan Sclaroff, Hanspeter Pfister
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We also develop methods for adapting our model to new subjects when a small number of subject-specific personalization data is available.  ...  We develop hierarchical Bayesian neural networks to capture subject-specific variations and share statistical strength across subjects.  ...  The mean F1-scores for different versions of our Hierarchical Bayesian gesture classifier.  ... 
doi:10.1109/cvpr.2017.56 dblp:conf/cvpr/JoshiGBSP17 fatcat:5k6dnxplrvftfh7ctelxv3sxte

Bayesian Performance Comparison of Text Classifiers

Dell Zhang, Jun Wang, Emine Yilmaz, Xiaoling Wang, Yuxin Zhou
2016 Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16  
to compare commonly-used multivariate performance measures like F1 scores instead of accuracy, and so on.  ...  In contrast to the existing probabilistic model for F1 scores which is unpaired, our proposed model takes the correlation between classifiers into account and thus achieves greater statistical power.  ...  outcomes, and they produce estimations for simple statistics (such as the average difference between the two given groups) but not complex performance measures (such as the F1 score).  ... 
doi:10.1145/2911451.2911547 dblp:conf/sigir/ZhangWYWZ16 fatcat:h642m4qr2zg3befxu2himexj2q

Automatic figure classification in bioscience literature

Daehyun Kim, Balaji Polepalli Ramesh, Hong Yu
2011 Journal of Biomedical Informatics  
The best system is a multi-model classifier which combines the rule-based hierarchical classifier and a support vector machine (SVM) based classifier, achieving a 76.7% F1-score for five-type classification  ...  We performed feature selection and explored different classification models, including a rule-based figure classifier, a supervised machine-learning classifier, and a multi-model classifier, the latter  ...  We thank Lamont Antieau and Feifan Liu for editing the article. We also thank Lisha Choubey for the inter-annotation of figure data.  ... 
doi:10.1016/j.jbi.2011.05.003 pmid:21645638 pmcid:PMC3176927 fatcat:gsacbtehibbntcgrm5j22jxtrm

Drift vs Shift: Decoupling Trends and Changepoint Analysis [article]

Haoxuan Wu, Sean Ryan, David S. Matteson
2022 arXiv   pre-print
Our locally adaptive model-based approach for robustly decoupling combines Bayesian trend filtering and machine learning based regularization.  ...  We introduce a new approach for decoupling trends (drift) and changepoints (shifts) in time series.  ...  DC-DS achieves a F1-score of 0.89 for setting of magnitude of change of 2, a F1-score of 0.62 for setting of magnitude of change of 1 and a F1-score of 0.33 for setting of magnitude of change of 0.5.  ... 
arXiv:2201.06606v2 fatcat:y6rmrhjgpvccnfqhaplr4qwmia
« Previous Showing results 1 — 15 out of 5,141 results