Publishing Bangladesh Socio-economic Data as Linked and Multidimensional Data

Inspired by the vision of achieving a ‘Smart Bangladesh,’ the Bangladesh Bureau of Statistics consistently endeavors to manage and disseminate socio-economic data on the web. However, the existing datasets face several challenges: 1) they are presented in an unsustainable, read-only, static, and aggregated format, 2) they lack adherence to the Findability, Accessibility, Interoperability, and Reusability (FAIR) principles, and 3) they are not conducive to the interactive analysis and integration with other data sources. In this paper, we propose a comprehensive solution by outlining a step-by-step process for constructing a knowledge graph specifically tailored to the Bangladesh census data. The resulting knowledge graph is presented using the Resource Description Framework (RDF) model, enriched with multidimensional semantics, facilitating Online Analytical Processing (OLAP)-like analysis and aligning with the FAIR and linked data principles. It has a size of 6.31 GB, is linked to 772 external resources, and contains approximately 39 million RDF triples. Our evaluation focuses on assessing both the integration process and the quality of the generated knowledge graph in terms of FAIR principles and OLAP operability.