A Causal Explanatory Model of Bayesian-belief Networks for Analysing the Risks of Opening Data

Luthfi, Ahmad; Janssen, Marijn; Crompvoets, Joep

doi:10.1007/978-3-319-94214-8_20

Cited by 10 publications

(15 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Iterative model of decision support for opening data [11,12] The use of Bayesian-belief networks approach is to construct the relational model of decision support in opening data. The outcomes of this model can be used to prevent the risks and still gain benefits of opening data.…”

Section: Methods Overview and Limitationsmentioning

confidence: 99%

Decision Tree Analysis for Estimating the Costs and Benefits of Disclosing Data

Luthfi

Janssen

Crompvoets

2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

The public expects government institutions to open their data to enable society to reap the benefits of these data. However, governments are often reluctant to disclose their data due to possible disadvantages. These disadvantages, at the same time, can be circumstances by processing the data before disclosing. Investments are needed to be able to pre-process a dataset. Hence, a trade-off between the benefits and cost of opening data needs to be made. Decisions to disclose are often made based on binary options like "open" or "closed" the data, whereas also parts of a dataset can be opened or only pre-processed data. The objective of this study is to develop a decision tree analysis in open data (DTOD) to estimate the costs and benefits of disclosing data using a DTA approach. Experts' judgment is used to quantify the pay-offs of possible consequences of the costs and benefits and to estimate the chance of occurrence. The result shows that for non-trivial decisions the DTOD helps, as it allows the creation of decision structures to show alternatives ways of opening data and the benefits and disadvantages of each alternative.

show abstract

Section: Methods Overview and Limitationsmentioning

confidence: 99%

Decision Tree Analysis for Estimating the Costs and Benefits of Disclosing Data

Luthfi

Janssen

Crompvoets

2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Many works of literature on Open Government Data (OGD) demonstrates that the disclosure of the OGD will enhance transparency and accountability of public administrations, realize the better governance, and generate confidence in the government's decisions [6] - [8]. Besides, some potential data users like researchers, businesses, or even non-public organizations can participate in evidencebased policy-making in the decision-making process [9].…”

Section: Introductionmentioning

confidence: 99%

“…Two main advantages are using evidence-based decision-making in open data initiatives. First, it can reduce uncertainty situation about which direction to follow, as the decision to release the data is based on objective evidence on which all parties are likely to agree on [8], [17]. Second, evidence policy-making has to base on reliable facts, and this can result in new insights on the policy issues to the public [15].…”

Section: Introductionmentioning

confidence: 99%

Open Data for Evidence-based Decision-making: Data-driven Government Resulting in Uncertainty and Polarization

Luthfi¹,

Janssen²

2019

Int. J. Adv. Sci. Eng. Inf. Technol.

Self Cite

View full text Add to dashboard Cite

Over the last decade, more and more data are collected and opened. Governments actively stimulate the opening of data to increase citizen engagement to support policy-making processes. Evidence-based policy-making is the situation whereby decisions made are based on factual data. The common expectation is that releasing data will result in evidence-based decision-making and more trust in government decisions. This study aims to provide insight into how evidence-based policy based on open data can result into uncertainty and even polarize the policy-making process. We analyze a case study in which traffic and road utilization datasets are used and model the decision-making process using the Business Process Model and Notation (BPMN). The BPMN model shows how the government and business organizations can use the data and give different interpretations. Data-driven decision-making might potentially create uncertainty, polarization, and less trust in decisions as stakeholders can give different meanings to the data and arrive at different outcomes. In contrast to the common belief, we found that the more data released, the more discussions happened about what is desired according to the data. The various directions derived from the data can even polarize decisionmaking. In other words, the more data opened, the more people can construct their perception of reality. For further research, we recommend understanding the types and role of data to create an evidence-based approach.

show abstract

“…At this moment, the governments may have their own business process to avoid human or technological system mistakes from open data decisions [5]. In reality, the way to analyze the risks, costs, and other effects-adverse of disclosing data to the potential stakeholders are cumbersome [6].…”

Section: Introductionmentioning

confidence: 99%

A Comparative Study of Methods for Deciding to Open Data

Luthfi

Janssen

2019

Lecture Notes in Business Information Processing

Self Cite

View full text Add to dashboard Cite

Governments may have their own business processes to decide to open data, which might be supported by decision-making tools. At the same time, analyzing potential benefits, costs, risks, and other effects-adverse of disclosing data are challenging. In the literature, there are various methods to analyze the potential advantages and disadvantages of opening data. Nevertheless, none of them provides discussion into the comparative studies in terms of strengths and weaknesses. In this study, we compare three methods for disclosing data, namely Bayesian-belief networks, Fuzzy multi-criteria decisionmaking, and Decision tree analysis. The comparative study is a mechanism for further studying the development of a knowledge domain by performing a feature-by-feature at the same level of functionalities. The result of this research shows that the methods have different strengths and weaknesses. The Bayesianbelief Networks has higher accuracy in comparison, and able to construct the causal relationships of the selected variable under uncertainties. Yet, this method is more resource intensive. This study can contribute to the decision-makers and respected researchers to a better comprehend and provide recommendation related to the three methods comparison.

show abstract

A Causal Explanatory Model of Bayesian-belief Networks for Analysing the Risks of Opening Data

Cited by 10 publications

References 20 publications

Decision Tree Analysis for Estimating the Costs and Benefits of Disclosing Data

Decision Tree Analysis for Estimating the Costs and Benefits of Disclosing Data

Open Data for Evidence-based Decision-making: Data-driven Government Resulting in Uncertainty and Polarization

A Comparative Study of Methods for Deciding to Open Data

Contact Info

Product

Resources

About