Linking Bayesian networks and PLS path modeling for causal analysis

Wu, Wei

doi:10.1016/j.eswa.2009.05.021

Cited by 71 publications

(45 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For example, in contrast to LISREL, PLS path modeling is more suitable for real applications. Particularly when models are more complex or the distribution of data is not-normal, the use of this approach would be more appropriate (43). The main advantage of PLS modeling compared to LISREL is that a smaller number of samples are required (44).…”

Section: Methodsmentioning

confidence: 99%

Development of the UTAUT2 model to measure the acceptance of medical laboratory portals by patients in Shiraz

Ravangard¹,

Kazemi²,

Abbasali³

et al. 2017

Electron physician

View full text Add to dashboard Cite

IntroductionOne of the main stages for achieving the success is acceptance of technology by its users. Hence, identifying the effective factors in successful acceptance of information technology is necessary and vital. One such factor is usability. This study aimed to investigate the software usability in the “Unified Theory of Acceptance and Use of Technology 2 (UTAUT2)” model in patients’ use of medical diagnosis laboratories’ electronic portals in 2015.MethodsThis cross-sectional study was carried out on 170 patients in 2015. A 27-item questionnaire adopted from previous research and the Usability Evaluation questionnaire were used for data collection. Data were analyzed using Structural Equation Modeling (SEM), with Partial Least Squares approach by SPSS 20.0 and Smart-PLS V3.0.ResultsThe results showed that the construct of intention to use had significant associations with price value (t-value=2.77), hedonic motivation (t-value=4.46), habit (t-value=1.99) and usability (t-value=5.2), as well as the construct of usage behavior with usability (t-value=3.45) and intention to use (t-value=2.03).ConclusionConsidering the results of this study, the following recommendations can be made in order for the higher use of portals by the patients: informing patients about the advantages of using these portals, designing portals in a simple and understandable form, increasing the portals’ attractiveness, etc.

show abstract

Section: Methodsmentioning

confidence: 99%

Development of the UTAUT2 model to measure the acceptance of medical laboratory portals by patients in Shiraz

Ravangard¹,

Kazemi²,

Abbasali³

et al. 2017

Electron physician

View full text Add to dashboard Cite

show abstract

“…The BCN has the advantage of having no rigid statistical assumptions. It graphically displays a directed acyclic graph and represents a set of conditional independence constraints among a given number of variables and their related conditional probability distributions (Wu, 2010). These BCNs can handle incomplete data sets and help easily model causal relationships to gain understanding about a problem domain and make predictions in the presence of interventions.…”

Section: Fundamentals Of Bcnsmentioning

confidence: 99%

“…Several search algorithms such as simulated annealing algorithms, genetic algorithms, and Three Augmented Naïve Bayes (TAN) algorithms (Cerquides and Mantaras, 2005) have been developed for this purpose (Wu, 2010;Hruschka and Ebecken, 2007;Baesens et al, 2004;). The knowledge-based approach, on the other hand, uses the causal knowledge of domain experts to construct networks.…”

Section: Fundamentals Of Bcnsmentioning

confidence: 99%

A decision support methodology to enhance the competitiveness of the Turkish automotive industry

Ülengin

Önsel

Aktaş

et al. 2014

European Journal of Operational Research

View full text Add to dashboard Cite

Three levels of competitiveness affect the success of business enterprises in a globally competitive environment: the competitiveness of the company, the competitiveness of the industry in which the company operates and the competitiveness of the country where the business is located. This study analyses the competitiveness of the automotive industry in association with the national competitiveness perspective using a methodology based on Bayesian Causal Networks. First, we structure the competitiveness problem of the automotive industry through a synthesis of expert knowledge in the light of the World Economic Forum's competitiveness indicators. Second, we model the relationships among the variables identified in the problem structuring stage and analyse these relationships using a Bayesian Causal Network. Third, we develop policy suggestions under various scenarios to enhance the national competitive advantages of the automotive industry. We present an analysis of the Turkish automotive industry as a case study. It is possible to generalise the policy suggestions developed for the case of Turkish automotive industry to the automotive industries in other developing countries where country and industry competitiveness levels are similar to those of Turkey.

show abstract

“…Moreover, they also provide good estimates even when some predictors are missing (Nicholson et al 2008, Nadkarni and. They are also more robust when compared to other methods while studying with data that contain noise (Wu 2010).…”

Section: Data and Methodsologymentioning

confidence: 98%

Does Gender Equality Lead to Better-Performing Economies A Bayesian Causal Map Approach

Yücel¹,

Ekici²

2017

dogus

View full text Add to dashboard Cite

This study explores the existence of relationships between gender inequalities -represented by the components of the World Economic Forum (WEF) Global Gender Gap Index-and the major macroeconomic indicators. The relationships within gender inequalities in education, the labour market, health and the political arena, and between gender inequalities and gross macroeconomic aggregates were modelled with the Bayesian Causal Map, an effective tool that is used to analyze cause-effect relations and conditional dependencies between variables. A data set of 128 countries during the period 2007-2011 is used. Findings reveal that some inequalities have high levels of interaction with each other. In addition, eradicating gender inequalities is found to be associated with better economic performance, mainly in the form of higher gross domestic product growth, investment, and competitiveness.

show abstract

Linking Bayesian networks and PLS path modeling for causal analysis

Cited by 71 publications

References 52 publications

Development of the UTAUT2 model to measure the acceptance of medical laboratory portals by patients in Shiraz

Development of the UTAUT2 model to measure the acceptance of medical laboratory portals by patients in Shiraz

A decision support methodology to enhance the competitiveness of the Turkish automotive industry

Does Gender Equality Lead to Better-Performing Economies A Bayesian Causal Map Approach

Contact Info

Product

Resources

About