Filters








343 Hits in 4.9 sec

Mining local and tail dependence structures based on pointwise mutual information

Teruko Takada
2011 Data mining and knowledge discovery  
The proposed pointwise mutual information estimator is sufficiently robust and efficient for exploring tail dependence, and its good performance was confirmed in an experimental study.  ...  This article proposes a novel approach for analyzing the entire structure of nonlinear dependence between two data sets on the basis of accurate pointwise mutual information estimation.  ...  Acknowledgements The author is deeply grateful to Roger Koenker and Shoji Takada for their many helpful suggestions and constructive comments.  ... 
doi:10.1007/s10618-011-0220-3 fatcat:gx3zsaqcubaufe2i77fvvag2ke

Erratum to: Mining local and tail dependence structures based on pointwise mutual information

Teruko Takada
2011 Data mining and knowledge discovery  
See Fig. 7 Fig. 8 78 Size effect on the pointwise MI estimation error of dependent bivariate t(3) density: Tail effect on the pointwise MI estimation of dependent bivariate t density: (Panels B) Pointwise  ...  See Fig. 9 9 Tail effect on the pointwise MI estimation error of dependent bivariate t density: Sample size n = 10000. See  ... 
doi:10.1007/s10618-011-0241-y fatcat:zgwzqapswzcopmjuetvcp7z6ku

Neural Motifs: Scene Graph Parsing with Global Context

Rowan Zellers, Mark Yatskar, Sam Thomson, Yejin Choi
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We present new quantitative insights on such repeated structures in the Visual Genome dataset. Our analysis shows that object labels are highly predictive of relation labels but not vice-versa.  ...  This baseline improves on the previous state-of-the-art by an average of 3.6% relative improvement across evaluation settings.  ...  We conducted an analysis of repeated motifs in scene graphs by mining combinations of object-relation-object labels that have high pointwise mutual information with each other.  ... 
doi:10.1109/cvpr.2018.00611 dblp:conf/cvpr/ZellersYTC18 fatcat:nvye4ywmyjdajpfei5zhqhn2lu

Neural Motifs: Scene Graph Parsing with Global Context [article]

Rowan Zellers, Mark Yatskar, Sam Thomson, Yejin Choi
2018 arXiv   pre-print
We present new quantitative insights on such repeated structures in the Visual Genome dataset. Our analysis shows that object labels are highly predictive of relation labels but not vice-versa.  ...  This baseline improves on the previous state-of-the-art by an average of 3.6% relative improvement across evaluation settings.  ...  We conducted an analysis of repeated motifs in scene graphs by mining combinations of object-relation-object labels that have high pointwise mutual information with each other.  ... 
arXiv:1711.06640v2 fatcat:hvuai4lihzhwxhaog674btnqn4

Parametric Copula-GP model for analyzing multidimensional neuronal and behavioral relationships [article]

Nina Kudryashova, Theoklitos Amvrosiadis, Nathalie Dupuy, Nathalie Rochefort, Arno Onken
2020 arXiv   pre-print
We validate the model on synthetic data and compare its performance in estimating mutual information against the commonly used non-parametric algorithms.  ...  We propose a parametric copula model which separates the statistics of the individual variables from their dependence structure, and escapes the curse of dimensionality by using vine copula constructions  ...  Acknowledgments and Disclosure of Funding  ... 
arXiv:2008.01007v1 fatcat:nwzxccgwq5alrcx7ovn2yrok6m

Fuzzy Students T-Distribution Model Based on Richer Spatial Combination

Tao Lei, Xiaohong Jia, Dinghua Xue, Qi Wang, Hongying Meng, Asoke K Nandi
2021 IEEE transactions on fuzzy systems  
The first is that both narrow and wide receptive fields are integrated into the objective function of FRSC, which is convenient to mine image features and distinguish local difference.  ...  Experimental results on synthetic and publicly available images, further demonstrate that the proposed FRSC addresses successfully the limitations of FCM algorithms with spatial information and provides  ...  For more information, see https://creativecommons.org/licenses/by/4.0/ This article has been accepted for publication in a future issue of this journal, but has not been fully edited.  ... 
doi:10.1109/tfuzz.2021.3099560 fatcat:zuczr4lnr5bg5pag3yknf7yn24

Relation Representation Learning Via Signed Graph Mutual Information Maximization for Trust Prediction

Yongjun Jing, Hao Wang, Kun Shao, Xing Huo
2021 Symmetry  
In SGMIM, we incorporate a translation model and positive point-wise mutual information to enhance the relation representations and adopt Mutual Information Maximization to align the entity and relation  ...  We conduct link sign prediction in trust networks based on learned the relation representation.  ...  Acknowledgments: The authors are grateful to the editor and referees for their valuable comments and suggestions for improving the paper.  ... 
doi:10.3390/sym13010115 fatcat:7fhxnc6dnbf2vha5lcvndclvtq

Discriminatory Target Learning: Mining Significant Dependence Relationships from Labeled and Unlabeled Data

Zhi-Yi Duan, Li-Min Wang, Musa Mammadov, Hua Lou, Ming-Hui Sun
2019 Entropy  
Most traditional BNCs tend to build only one model to fit training instances by analyzing independence between attributes using conditional mutual information.  ...  to state-of-the-art learners such as Random forest and averaged one-dependence estimators.  ...  For unlabeled data point x = {x 1 , · · · , x n }, the conditional dependence between X i and X j (1 ≤ i, j ≤ n) with respect to label y on point x is measured by pointwise y-conditional mutual information  ... 
doi:10.3390/e21050537 pmid:33267251 fatcat:z7tvblff2zdbldg3g4cwtxm3ja

Bayesian probabilistic models for corporate context, with an application to internal audit activities [article]

Francesco Toraldo, Fabio S. Priuli
2021 arXiv   pre-print
We thus present the different stages of such analytical strategy, from feature selection, to model structure inference and model selection, as a general toolbox that allows a completely transparent and  ...  Phase 1 -variable selection based on the concepts of mutual information and conditional mutual information; • Phase 2 -structure inference of Bayesian networks for the process of interest, combinining  ...  Variable selection The first step consists of selecting the most promising predictors, based on mutual information scores (both unconditional and conditional) as defined in (5) .  ... 
arXiv:2007.06084v2 fatcat:hx6764vy7rcghatnh6stftfinm

Synonym Discovery for Structured Entities on Heterogeneous Graphs

Xiang Ren, Tao Cheng
2015 Proceedings of the 24th International Conference on World Wide Web - WWW '15 Companion  
Unlike existing query logbased methods, we delve deeper to explore sub-queries, and exploit tailed synonyms and tailed web pages for harvesting more synonyms.  ...  A general, heterogeneous graph-based data model which encodes our problem insights is designed by capturing three key concepts (synonym candidate, web page and keyword) and different types of interactions  ...  Specifically, we extend Pointwise Mutual Information (PMI) [30] to measure the collocation for n-grams in N .  ... 
doi:10.1145/2740908.2745396 dblp:conf/www/RenC15 fatcat:3h7h77zdozdx3noz53yjqq5zem

Locality and compositionality in zero-shot learning [article]

Tristan Sylvain, Linda Petrini, Devon Hjelm
2019 arXiv   pre-print
, are both deeply related to generalization and motivate the focus on more local-aware models in future research directions for representation learning.  ...  In this work we study locality and compositionality in the context of learning representations for Zero Shot Learning (ZSL).  ...  on Mutual Information, that allows to investigate local properties of the learned representations.  ... 
arXiv:1912.12179v1 fatcat:ewkdisldxrgupb2grshj4ucncm

A Review on Algorithms for Constraint-based Causal Discovery [article]

Kui Yu, Jiuyong Li, Lin Liu
2016 arXiv   pre-print
Recent years, as the availability of abundant large-sized and complex observational data, the constrain-based approaches have gradually attracted a lot of interest and have been widely applied to many  ...  Secondly and primarily, the state-of-the-art constraint-based casual inference algorithms are surveyed with the detailed analysis.  ...  • Step 4: Orienting the v-structures again based on the updated skeleton and the updated information in sepset.  ... 
arXiv:1611.03977v2 fatcat:ercpfkqssnabfgdc3ndd7bd3tu

Discriminative Structure Learning of Bayesian Network Classifiers from Training Dataset and Testing Instance

LiMin Wang, Yang Liu, Musa Mammadov, MingHui Sun, Sikai Qi
2019 Entropy  
To reduce the search space of possible attribute orders, k-dependence Bayesian classifier (KDB) simply applies mutual information to sort attributes.  ...  The extensive experimental results on the Wisconsin breast cancer database for case study and other 10 datasets by involving classifiers with different structure complexities, such as Naive Bayes (0-dependence  ...  Since the efficiency of the UKDB depends on the efficiency of MI and CMI, we use another criterion, pointwise mutual information (PMI) and pointwise conditional mutual information (PCMI) to compare and  ... 
doi:10.3390/e21050489 pmid:33267204 pmcid:PMC7514978 fatcat:qdbdcdbvjngcdehbx3fup222hq

KnowAugNet: Multi-Source Medical Knowledge Augmented Medication Prediction Network with Multi-Level Graph Contrastive Learning [article]

Yang An, Bo Jin, Xiaopeng Wei
2022 arXiv   pre-print
Most existing studies focus on utilizing inherent relations between homogeneous codes of medical ontology graph to enhance their representations using supervised methods, and few studies pay attention  ...  However, medication prediction is a challenging data mining task due to the complex relations between medical codes.  ...  Acknowledgements Funding: This research was partially supported by the National Key R&D Program of China (2018YFC0116800), National Natural Science Foundation of China (No. 61772110, 6217072188 and 71901011  ... 
arXiv:2204.11736v2 fatcat:vvncqo2tejgwrbpo2cxpstqdt4

Verbal Characterization of Probabilistic Clusters using Minimal Discriminative Propositions [article]

Yoshitaka Kameya, Satoru Nakamura, Tatsuya Iwasaki, Taisuke Sato
2011 arXiv   pre-print
In a knowledge discovery process, interpretation and evaluation of the mined results are indispensable in practice.  ...  The proposed method provides us with a new, in-depth and consistent tool for cluster interpretation/evaluation, and works for various types of datasets including continuous attributes and missing values  ...  The authors would like to thank Toshihiro Kamishima for his helpful comments on related work.  ... 
arXiv:1108.5002v2 fatcat:mxu2olge75cvniqp25tiiqp7ba
« Previous Showing results 1 — 15 out of 343 results