The next generation of Open Government Data Platform (OGD+)

Open Government Data (OGD) can be a powerful tool for Small and Medium-sized Enterprises (SMEs), helping them make better decisions, increase efficiency, and gain a competitive advantage. By analysing large and complex datasets, SMEs can benefit significantly from utilising OGD in improved decision-making. Further, OGD can help SMEs automate processes, streamline operations, and reduce waste, increasing efficiency and productivity. Despite the benefits of OGD, several challenges are associated with its use, such as quality control, data accessibility, data security and privacy, data interpretation, resource constraints, and data representation.

Moreover, the current OGD frameworks need help to address all these challenges. They need to provide modern data analytics tools that use Artificial Intelligence (AI). Thus, this thesis presents the next generation of OGD frameworks that is to be used by SMEs. The first part developed an approach named the consolidated criteria for evaluating OGD initiatives derived from the literature review and the desired features of OGD. Moreover, the tight criteria for evaluating OGD initiatives proved that most OGD initiatives still need to be developed in data analytics and the benefits that machine learning may provide when incorporated into OGD frameworks, such as simplifying operations, informing business decisions, and reducing risks.

The second part provides an analysis and the key findings of each data collection method to find and analyse research gaps. The research gap analysis summarises the key conclusions that combine the issues with existing OGD frameworks and the required features from the study of conducted data collection methods. These findings and other requirements were used to propose the OGD+, the next generation of the OGD framework. Dashboard, Query Editor, Schema, and Data are the four layers of the OGD+ analytics service.

The proposed OGD+ was assessed to investigate its features regarding ML algorithms, their correctness, and how easily users can access and query datasets on the proposed OGD+. In addition, the accuracy and results of the ML algorithms demonstrated that combining Bert and Belts ML algorithms will increase sentiment analysis findings by at least 1%. Fifty SMEs took part in the user experience survey. The data gathered indicated a roughly 94% positive appraisal of the proposed OGD+. Furthermore, all participants were satisfied with the proposed OGD+ concepts, features, or data.