The Experts below are selected from a list of 12828 Experts worldwide ranked by ideXlab platform
Hendrik Blockeel - One of the best experts on this subject based on the ideXlab platform.
-
Contextual Variable elimination with overlapping contexts
Probabilistic Graphical Models, 2010Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:Belief networks (BNs) extracted from statistical relational learning formalisms often include Variables with conditional probability distributions (CPDs) that exhibit a local structure (e.g, decision trees and noisy-or). In such cases, naively representing CPDs as tables and using a general purpose inference algorithm such as Variable elimination (VE) results in redundant computation. Contextual Variable elimination (CVE) partly addresses this problem by representing the BN in terms of smaller units called confactors. This leads to a more compact representation and faster inference. CVE requires that a Variable’s confactors are mutually-exclusive and exhaustive. We propose CVE-OC (CVE with overlapping contexts), which lifts these restrictions. This seemingly simple step shows to be powerful and allows for a more efficient encoding of confactors and a reduction of the computational cost. Experiments show that CVE-OC outperforms CVE on multiple problems.
-
cp logic theory inference with Contextual Variable elimination and comparison to bdd based inference methods
Inductive Logic Programming, 2009Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:There is a growing interest in languages that combine probabilistic models with logic to represent complex domains involving uncertainty. Causal probabilistic logic (CP-logic), which has been designed to model causal processes, is such a probabilistic logic language. This paper investigates inference algorithms for CP-logic; these are crucial for developing learning algorithms. It proposes a new CP-logic inference method based on Contextual Variable elimination and compares this method to Variable elimination and to methods based on binary decision diagrams.
-
ILP - CP-logic theory inference with Contextual Variable elimination and comparison to BDD based inference methods
Inductive Logic Programming, 2009Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:There is a growing interest in languages that combine probabilistic models with logic to represent complex domains involving uncertainty. Causal probabilistic logic (CP-logic), which has been designed to model causal processes, is such a probabilistic logic language. This paper investigates inference algorithms for CP-logic; these are crucial for developing learning algorithms. It proposes a new CP-logic inference method based on Contextual Variable elimination and compares this method to Variable elimination and to methods based on binary decision diagrams.
Wannes Meert - One of the best experts on this subject based on the ideXlab platform.
-
Contextual Variable elimination with overlapping contexts
Probabilistic Graphical Models, 2010Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:Belief networks (BNs) extracted from statistical relational learning formalisms often include Variables with conditional probability distributions (CPDs) that exhibit a local structure (e.g, decision trees and noisy-or). In such cases, naively representing CPDs as tables and using a general purpose inference algorithm such as Variable elimination (VE) results in redundant computation. Contextual Variable elimination (CVE) partly addresses this problem by representing the BN in terms of smaller units called confactors. This leads to a more compact representation and faster inference. CVE requires that a Variable’s confactors are mutually-exclusive and exhaustive. We propose CVE-OC (CVE with overlapping contexts), which lifts these restrictions. This seemingly simple step shows to be powerful and allows for a more efficient encoding of confactors and a reduction of the computational cost. Experiments show that CVE-OC outperforms CVE on multiple problems.
-
cp logic theory inference with Contextual Variable elimination and comparison to bdd based inference methods
Inductive Logic Programming, 2009Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:There is a growing interest in languages that combine probabilistic models with logic to represent complex domains involving uncertainty. Causal probabilistic logic (CP-logic), which has been designed to model causal processes, is such a probabilistic logic language. This paper investigates inference algorithms for CP-logic; these are crucial for developing learning algorithms. It proposes a new CP-logic inference method based on Contextual Variable elimination and compares this method to Variable elimination and to methods based on binary decision diagrams.
-
ILP - CP-logic theory inference with Contextual Variable elimination and comparison to BDD based inference methods
Inductive Logic Programming, 2009Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:There is a growing interest in languages that combine probabilistic models with logic to represent complex domains involving uncertainty. Causal probabilistic logic (CP-logic), which has been designed to model causal processes, is such a probabilistic logic language. This paper investigates inference algorithms for CP-logic; these are crucial for developing learning algorithms. It proposes a new CP-logic inference method based on Contextual Variable elimination and compares this method to Variable elimination and to methods based on binary decision diagrams.
Jan Struyf - One of the best experts on this subject based on the ideXlab platform.
-
Contextual Variable elimination with overlapping contexts
Probabilistic Graphical Models, 2010Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:Belief networks (BNs) extracted from statistical relational learning formalisms often include Variables with conditional probability distributions (CPDs) that exhibit a local structure (e.g, decision trees and noisy-or). In such cases, naively representing CPDs as tables and using a general purpose inference algorithm such as Variable elimination (VE) results in redundant computation. Contextual Variable elimination (CVE) partly addresses this problem by representing the BN in terms of smaller units called confactors. This leads to a more compact representation and faster inference. CVE requires that a Variable’s confactors are mutually-exclusive and exhaustive. We propose CVE-OC (CVE with overlapping contexts), which lifts these restrictions. This seemingly simple step shows to be powerful and allows for a more efficient encoding of confactors and a reduction of the computational cost. Experiments show that CVE-OC outperforms CVE on multiple problems.
-
cp logic theory inference with Contextual Variable elimination and comparison to bdd based inference methods
Inductive Logic Programming, 2009Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:There is a growing interest in languages that combine probabilistic models with logic to represent complex domains involving uncertainty. Causal probabilistic logic (CP-logic), which has been designed to model causal processes, is such a probabilistic logic language. This paper investigates inference algorithms for CP-logic; these are crucial for developing learning algorithms. It proposes a new CP-logic inference method based on Contextual Variable elimination and compares this method to Variable elimination and to methods based on binary decision diagrams.
-
ILP - CP-logic theory inference with Contextual Variable elimination and comparison to BDD based inference methods
Inductive Logic Programming, 2009Co-Authors: Wannes Meert, Jan Struyf, Hendrik BlockeelAbstract:There is a growing interest in languages that combine probabilistic models with logic to represent complex domains involving uncertainty. Causal probabilistic logic (CP-logic), which has been designed to model causal processes, is such a probabilistic logic language. This paper investigates inference algorithms for CP-logic; these are crucial for developing learning algorithms. It proposes a new CP-logic inference method based on Contextual Variable elimination and compares this method to Variable elimination and to methods based on binary decision diagrams.
Pierre Chauvin - One of the best experts on this subject based on the ideXlab platform.
-
Which adults in the Paris metropolitan area have never been tested for HIV? A 2010 multilevel, cross-sectional, population-based study.
BMC Infectious Diseases, 2015Co-Authors: Véronique Massari, Annabelle Lapostolle, Marie-catherine Grupposo, Rosemary Dray-spira, Dominique Costagliola, Pierre ChauvinAbstract:Despite the widespread offer of free HIV testing in France, the proportion of people who have never been tested remains high. The objective of this study was to identify, in men and women separately, the various factors independently associated with no lifetime HIV testing. We used multilevel logistic regression models on data from the SIRS cohort, which included 3006 French-speaking adults as a representative sample of the adult population in the Paris metropolitan area in 2010. The lifetime absence of any HIV testing was studied in relation to individual demographic and socioeconomic factors, psychosocial characteristics, sexual biographies, HIV prevention behaviors, attitudes towards people living with HIV/AIDS (PLWHA), and certain neighborhood characteristics. In 2010, in the Paris area, men were less likely to have been tested for HIV at least once during their lifetime than women. In multivariate analysis, in both sexes, never having been tested was significantly associated with an age younger or older than the middle-age group (30-44 years), a low education level, a low self-perception of HIV risk, not knowing any PLWHA, a low lifetime number of couple relationships, and the absence of any history of STIs. In women, other associated factors were not having a child < 20 years of age, not having additional health insurance, having had no or only one sexual partner in the previous 5 years, living in a cohabiting couple or having no relationship at the time of the survey, and a feeling of belonging to a community. Men with specific health insurance for low-income individuals were less likely to have never been tested, and those with a high stigma score towards PLWHA were more likely to be never-testers. Our study also found neighborhood differences in the likelihood of men never having been tested, which was, at least partially, explained by the neighborhood proportion of immigrants. In contrast, in women, no Contextual Variable was significantly associated with never-testing for HIV after adjustment for individual characteristics. Studies such as this one can help target people who have never been tested in the context of recommendations for universal HIV screening in primary care.
-
Investigating the spatial variability in incidence of coronary heart disease in the Gazel cohort: the impact of area socioeconomic position and mediating role of risk factors.
Journal of Epidemiology and Community Health, 2011Co-Authors: Romain Silhol, Marie Zins, Pierre Chauvin, Basile ChaixAbstract:STUDY OBJECTIVE: The aim of the study was to improve understanding of the relationships between Contextual socioeconomic characteristics and coronary heart disease (CHD) incidence in France. Several authors have suggested that CHD risk factors (diabetes, hypertension, cholesterol, overweight, tobacco consumption) may partly mediate associations between socioeconomic environmental Variables and CHD. Studies have assessed the overall mediating role of CHD risk factors, but have never investigated the specific mediating role of each risk factor, not allowing their specific contribution to the area socioeconomic position-CHD association to be disentangled. DESIGN: After assessing geographical variations in CHD incidence and socioeconomic environmental effects on CHD using a multilevel Cox model, the extent to which this Contextual effect was mediated by each of the CHD risk factors was assessed. PARTICIPANTS: Data of the French GAZEL cohort (n=19,808) were used. MAIN RESULTS: After adjustment for several individual socioeconomic indicators, it was found, in men from highly urbanised environments, that CHD incidence increased with decreasing socioeconomic position of the residential environment. After individual-level adjustment, a higher risk of obesity, smoking and cholesterol was observed in the most deprived residential environments. When risk factors were introduced into the model, a modest decrease was observed in the magnitude of the association between the socioeconomic Contextual Variable and CHD. Risk factors that contributed most to the decrease of the association were smoking and cholesterol. CONCLUSIONS: Classic risk factors, although some of them more than others, mediated a modest part of the association between area socioeconomic position and CHD.
Andrew Rundle - One of the best experts on this subject based on the ideXlab platform.
-
there goes the neighborhood effect bias owing to nondifferential measurement error in the construction of neighborhood Contextual measures
Epidemiology, 2014Co-Authors: Stephen J Mooney, Catherine Richards, Andrew RundleAbstract:Within epidemiology there is a large literature on the effects of non-differential measurement error or misclassification of individuals’ exposures on the magnitude of statistical association: such error usually biases an epidemiological effect estimate to the null1. For instance, in a study of the effect of personal income on body mass index (BMI) (e.g. 2), non-differential measurement error in income assessment would decrease the apparent association between income and BMI. There is a smaller literature on the effects of exposure measurement error in cross-sectional ecologic studies, such as a study relating the proportion of residents living below poverty across counties to the proportion of people in those counties who are obese3,4. For this type of study, data on individual poverty status is aggregated to determine the proportion of residents in each county who live in poverty; if there is misclassification of individuals’ poverty status, effect estimates are biased away from the null4. However, there is almost no literature on how non-differential measurement error or misclassification of exposure affects multilevel studies of neighborhood health effects on individual’s health outcomes. Multilevel neighborhood health effect studies typically include individual level outcome measures in a study population (e.g. individual’s BMI), individual level predictors measured from the study participants (e.g. age, race/ethnicity, gender, income) and neighborhood level Contextual predictors derived from Census data (e.g. neighborhood poverty rates). One of the goals of such studies is to quantify the effect of a neighborhood context feature (e.g. neighborhood poverty) on individual outcomes after adjusting for individual covariates. Neighborhood Contextual measures are often computed by aggregating data collected from individuals during the Decennial Census, American Community Survey, or another large social survey. Sensitive information, including income, is likely to be measured with error in these surveys5. A detailed review has suggested that under-reporting and over-reporting are about equally common6; for many outcomes, researchers might reasonably expect such error to be non-differential. It has previously been argued that non-differential measurement error in survey data causes bias away from the null for effect estimates relating outcome Variables measured at the individual level in a study population to a neighborhood Contextual Variable created by aggregating individual-level survey data from respondents who live in each neighborhood7. However, the potential magnitude of the bias has not been documented, nor has the quantitative relationship between the extent of measurement error and bias. For a given neighborhood context measure created by aggregating data collected from individuals, an investigator may have several options as to how to create a Contextual Variable8. For instance, ‘median household income’ and ‘proportion of residents living in poverty’ are both commonly used neighborhood Contextual Variables derived from the same individual-level Census data on income. Similarly, the investigator has several options for categorizing neighborhoods by a Contextual Variable9. For example, an investigator may choose to compare neighborhoods with less than 5% of residents living in poverty with neighborhoods where more than 20% of residents live in poverty or may choose to compare highest quintile to the lowest quintile of neighborhoods ranked by the percent of residents living in poverty. The effects of these choices on the extent of bias in calculated effect estimates when the underlying data includes measurement error have not been documented. Here we demonstrate the bias that occurs in the effect estimate for the influence of neighborhood Contextual Variables on individual level outcomes when there is measurement error or misclassification in the individual level data that is used to create the Contextual Variable. We demonstrate how choices an investigator makes in creation of Contextual Variables, such as the use of proportions as opposed to means, impact bias in the effect estimate, and how further manipulating the data to express continuous measures as categories affects bias.