A semantic network analysis of categorization in open government data portals

Abstract

Purpose

This study aims to evaluate the semantic relationships between category terms that are used in open government data (OGD) portals and those identified in policy documents through the implementation of a semantic network analysis.

Design/methodology/approach

This study was conducted in three stages. Firstly, the study examined the semantic relationships between category terms in OGD portals by constructing a similarity matrix based on the terms’ co-occurrence and visualizing six-word groups. Secondly, the study investigated the semantic relationships among terms in OGD policy documents using latent semantic analysis and community detection methods, resulting in the identification and visualization of three network groups. Finally, the study used chi-squared and Z-tests to analyse differences in category terms between countries with and without redefined categories.

Findings

The results indicate that the three-word groups were identified by community detection, covering various aspects of government. In addition, there is a significant difference between the two country groups, with category terms being more prevalent in countries with predefined categories. This emphasizes the impact of categorization on term prevalence within OGD portals.

Originality/value

This study uniquely focuses on the categorization of government portals for sustainable open data management. The findings underscore the importance of effectively structuring and organizing data categories to enhance user discoverability and accessibility in OGD portals.

Citation

Park, E.G. (2025), "A semantic network analysis of categorization in open government data portals", The Electronic Library, Vol. 43 No. 1, pp. 41-60. https://doi.org/10.1108/EL-05-2024-0147