Constraint-based causal discovery with mixed data

Tsagris, Michail; Borboudakis, Giorgos; Lagani, Vincenzo; Tsamardinos, Ioannis

doi:10.1007/s41060-018-0097-y

Cited by 34 publications

(22 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We first tested the mixed-type data extension of MIIC network reconstruction method on benchmark mixed-type data. Datasets were generated based on non-linear bayesian rules using the R script provided as Supplementary code; an example of non-Gaussian mixed-type distribution dataset is shown in S10 [25], also designed to analyze mixed-type data. Precision, Comparisons with fully continuous datasets, S13 Fig, were also performed with additional methods, CAM [26], kPC, rank-PC and rank-FCI [27] algorithms, S14 and S15 Figs, and confirm the better performance of MIIC over alternative continuous or mixed-type network learning methods.…”

Section: Application To Benchmark Synthetic Datamentioning

confidence: 99%

“…Another confounding example, X Z ! Y, taken from [25] with a uniform categorical Z with three levels, X and Y being continuous, for N = 1, 000 observations. With Z i the binary variable corresponding to the i-th dummy variable of Z, we defined X = −Z 1 + Z 2 + 0.2� X which is centered around either −1 if Z = 1, 0 if Z = 3 or 1 if Z = 2; and Y = Z 1 + Z 2 + 0.2� Y , � � N ð0; 1Þ which is centered around either 0 if Z = 3 or 1 if Z = 1 or Z = 2.…”

Section: S8 Fig Pairwise Dependence and Conditional Independence Betmentioning

confidence: 99%

See 1 more Smart Citation

Learning clinical networks from medical records based on information estimates in mixed-type data

et al. 2020

View full text Add to dashboard Cite

The precise diagnostics of complex diseases require to integrate a large amount of information from heterogeneous clinical and biomedical data, whose direct and indirect interdependences are notoriously difficult to assess. To this end, we propose an efficient computational approach to simultaneously compute and assess the significance of multivariate information between any combination of mixed-type (continuous/categorical) variables. The method is then used to uncover direct, indirect and possibly causal relationships between mixed-type data from medical records, by extending a recent machine learning method to reconstruct graphical models beyond simple categorical datasets. The method is shown to outperform existing tools on benchmark mixed-type datasets, before being applied to analyze the medical records of eldery patients with cognitive disorders from La Pitié-Salpêtrière Hospital, Paris. The resulting clinical network visually captures the global interdependences in these medical records and some facets of clinical diagnosis practice, without specific hypothesis nor prior knowledge on any clinically relevant information. In particular, it provides some physiological insights linking the consequence of cerebrovascular accidents to the atrophy of important brain structures associated to cognitive impairment.

show abstract

Section: Application To Benchmark Synthetic Datamentioning

confidence: 99%

Section: S8 Fig Pairwise Dependence and Conditional Independence Betmentioning

confidence: 99%

Learning clinical networks from medical records based on information estimates in mixed-type data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Meanwhile, there has also been more and more evidence demonstrating the possibility of discovering causal relationships by combining both experimental and observational data (Cooper and Yoo, 2013 ; Hauser and Bühlmann, 2015 ; Meinshausen et al, 2016 ). Other notable direction involves learning from mixed data types (continuous and discrete variables) (Andrews et al, 2018 ; Tsagris et al, 2018 ). In principle, our approach can be naturally adapted to handle mixed variable types, as long as an appropriate conditional independence test is employed.…”

Section: Background and Related Workmentioning

confidence: 99%

Causal Learning From Predictive Modeling for Observational Data

Ramanan

Natarajan

2020

Front. Big Data

View full text Add to dashboard Cite

We consider the problem of learning structured causal models from observational data. In this work, we use causal Bayesian networks to represent causal relationships among model variables. To this effect, we explore the use of two types of independencies-context-specific independence (CSI) and mutual independence (MI). We use CSI to identify the candidate set of causal relationships and then use MI to quantify their strengths and construct a causal model. We validate the learned models on benchmark networks and demonstrate the effectiveness when compared to some of the state-of-the-art Causal Bayesian Network Learning algorithms from observational Data.

show abstract

“…We also extend this method to make it applicable on heterogeneous data. Handling this type of data is a frequent issue in practice, and constitutes an active area of research [12,13].…”

Section: Introductionmentioning

confidence: 99%

Unsupervised extra trees: a stochastic approach to compute similarities in heterogeneous data

Dalleau

Couceiro

Smaïl-Tabbone

2020

Int J Data Sci Anal

View full text Add to dashboard Cite

In this paper we present a method to compute similarities on unlabeled data, based on extremely randomized trees. The main idea of our method, Unsupervised Extremely Randomized Trees (UET) is to randomly split the data in an iterative fashion until a stopping criterion is met, and to compute a similarity based on the co-occurrence of samples in the leaves of each generated tree. Using a tree-based approach to compute similarities is interesting, as the inherent We evaluate our method on synthetic and real-world datasets by comparing the mean similarities between samples with the same label and the mean similarities between samples with different labels. These metrics are similar to intracluster and intercluster similarities, and are used to assess the computed similarities instead of a clustering algorithm's results. Our empirical study shows that the method effectively gives distinct similarity values between samples belonging to different clusters, and gives indiscernible values when there is no cluster structure. We also assess some interesting properties such as invariance under monotone transformations of variables and robustness to correlated variables and noise. Finally, we performed hierarchical agglomerative clustering on synthetic and real-world homogeneous and heterogeneous datasets using UET versus standard simi-This paper is an extended version of the PAKDD'2018 Long Presentation paper "Unsupervised Extremely Randomized Trees".

show abstract

Constraint-based causal discovery with mixed data

Cited by 34 publications

References 38 publications

Learning clinical networks from medical records based on information estimates in mixed-type data

Learning clinical networks from medical records based on information estimates in mixed-type data

Causal Learning From Predictive Modeling for Observational Data

Unsupervised extra trees: a stochastic approach to compute similarities in heterogeneous data

Contact Info

Product

Resources

About