An FCA-based Approach to Direct Edges in a Causal Bayesian Network: A Pilot Study using a Surgery Data Set

Author(s):  
Walisson Ferreira ◽  
Mark Song ◽  
Luis Zarate
1998 ◽  
Vol 14 (3) ◽  
pp. 202-210 ◽  
Author(s):  
Suzanne Skiffington ◽  
Ephrem Fernandez ◽  
Ken McFarland

This study extends previous attempts to assess emotion with single adjective descriptors, by examining semantic as well as cognitive, motivational, and intensity features of emotions. The focus was on seven negative emotions common to several emotion typologies: anger, fear, sadness, shame, pity, jealousy, and contempt. For each of these emotions, seven items were generated corresponding to cognitive appraisal about the self, cognitive appraisal about the environment, action tendency, action fantasy, synonym, antonym, and intensity range of the emotion, respectively. A pilot study established that 48 of the 49 items were linked predominantly to the specific emotions as predicted. The main data set comprising 700 subjects' ratings of relatedness between items and emotions was subjected to a series of factor analyses, which revealed that 44 of the 49 items loaded on the emotion constructs as predicted. A final factor analysis of these items uncovered seven factors accounting for 39% of the variance. These emergent factors corresponded to the hypothesized emotion constructs, with the exception of anger and fear, which were somewhat confounded. These findings lay the groundwork for the construction of an instrument to assess emotions multicomponentially.


2018 ◽  
Vol 27 (4) ◽  
pp. 191-198
Author(s):  
Karen Van den Bussche ◽  
Sofie Verhaeghe ◽  
Ann Van Hecke ◽  
Dimitri Beeckman

2020 ◽  
pp. 003329412097815
Author(s):  
Giovanni Briganti ◽  
Donald R. Williams ◽  
Joris Mulder ◽  
Paul Linkowski

The aim of this work is to explore the construct of autistic traits through the lens of network analysis with recently introduced Bayesian methods. A conditional dependence network structure was estimated from a data set composed of 649 university students that completed an autistic traits questionnaire. The connectedness of the network is also explored, as well as sex differences among female and male subjects in regard to network connectivity. The strongest connections in the network are found between items that measure similar autistic traits. Traits related to social skills are the most interconnected items in the network. Sex differences are found between female and male subjects. The Bayesian network analysis offers new insight on the connectivity of autistic traits as well as confirms several findings in the autism literature.


2021 ◽  
Vol 27 (3) ◽  
pp. 8-34
Author(s):  
Tatyana Cherkashina

The article presents the experience of converting non-targeted administrative data into research data, using as an example data on the income and property of deputies from local legislative bodies of the Russian Federation for 2019, collected as part of anticorruption operations. This particular empirical fragment was selected for the pilot study of administrative data, which includes assessing the possibility of integrating scattered fragments of information into a single database, assessing quality of data and their relevance for solving research problems, particularly analysis of high-income strata and the apparent trends towards individualization of private property. The system of indicators for assessing data quality includes their timeliness, availability, interpretability, reliability, comparability, coherence, errors of representation and measurement, and relevance. In the case of the data set in question, measurement errors are more common than representation errors. Overall the article emphasizes the notion that introducing new non-target data into circulation requires their preliminary testing, while data quality assessment becomes distributed both in time and between different subjects. The transition from created data to «obtained» data shifts the functions of evaluating its quality from the researcher-creator to the researcheruser. And though in this case data quality is in part ensured by the legal support for their production, the transformation of administrative data into research data involves assessing a variety of quality measurements — from availability to uniformity and accuracy.


2015 ◽  
Vol 33 ◽  
pp. 82-90 ◽  
Author(s):  
Alan M. Langford ◽  
Jennifer R. Bolton ◽  
Michelle G. Carlin ◽  
Ray Palmer

Author(s):  
Jing ("Jim") Quan

This study examines influencing factors for users' intentions to tap through mobile advertisements. This chapter uses a data set with 115,899 records of ad tap-through from a mobile advertising company in China to fit a logit model to examine how the probability of advertisement tap-through is related to the identified factors. The results show that the influencing variables are application type, mobile operators, scrolling frequency, and the regional income level as they are positively correlated with the likelihood whether users would tap on certain types of advertising. Moreover, a Bayesian network model is used to estimate the conditional probability for a user to tap on an advertisement in an application after the user already taps on another advertisement in the same application. Based on the findings, strategies for mobile advertisers to engage in effective and targeted mobile advertising are proposed.


HLA ◽  
2020 ◽  
Vol 96 (2) ◽  
pp. 192-193
Author(s):  
Edwina Sutton ◽  
Dianne De Santis ◽  
Louise Hay ◽  
Elizabeth McKinnon ◽  
Lloyd D'Orsogna ◽  
...  

2018 ◽  
Vol 3 (4) ◽  
pp. 4046-4053
Author(s):  
Sujee Lee ◽  
Sijie Wang ◽  
Philip A. Bain ◽  
Christine Baker ◽  
Tammy Kundinger ◽  
...  

Author(s):  
Leonid Gutkin ◽  
Suresh Datla ◽  
Christopher Manu

Canadian Nuclear Standard CSA N285.8, “Technical requirements for in-service evaluation of zirconium alloy pressure tubes in CANDU® reactors”(1), permits the use of probabilistic methods when assessments of the reactor core are performed. A non-mandatory annex has been proposed for inclusion in the CSA Standard N285.8 to provide guidelines for performing uncertainty analysis in probabilistic fitness-for-service evaluations within the scope of this Standard, such as the probabilistic evaluation of leak-before-break. The proposed annex outlines the general approach to uncertainty analysis as being comprised of the following major activities: identification of influential variables, characterization of uncertainties in influential variables, and subsequent propagation of these uncertainties through the evaluation framework or code. The proposed methodology distinguishes between two types of non-deterministic variables by the method used to obtain their best estimate. Uncertainties are classified by their source, and different uncertainty components are considered when the best estimates for the variables of interest are obtained using calibrated parametric models or analyses and when these estimates are obtained using statistical models or analyses. The application of the proposed guidelines for uncertainty analysis was exercised by performing a pilot study for one of the evaluations within the scope of the CSA Standard N285.8, the probabilistic evaluation of leak-before-break based on a postulated through-wall crack. The pilot study was performed for a representative CANDU reactor unit using the recently developed software code P-LBB that complies with the requirements of Canadian Nuclear Standard CSA N286.7 for quality assurance of analytical, scientific, and design computer programs for nuclear power plants. This paper discusses the approaches used and the results obtained in the second stage of this pilot study, the uncertainty characterization of influential variables identified as discussed in the companion paper presented at the PVP 2018 Conference (PVP2018-85010). In the proposed methodology, statistical assessment and expert judgment are recognized as two complementary approaches to uncertainty characterization. In this pilot study, the uncertainty characterization was limited to cases where statistical assessment could be used as the primary approach. Parametric uncertainty and uncertainty due to numerical solutions were considered as the uncertainty components for variables represented by parametric models. Residual uncertainty and uncertainty due to imbalances in the model-basis data set were considered as the uncertainty components for variables represented by statistical models. In general, the uncertainty due to numerical solutions was found to be substantially smaller than the parametric uncertainty for variables represented by parametric models, and the uncertainty due to imbalances in the model basis data set was found to be substantially smaller than the residual uncertainty for variables represented by statistical models.


Sign in / Sign up

Export Citation Format

Share Document