Rethinking Thematic Evolution in Science Mapping: An Integrated Framework for Longitudinal Analysis

TL;DR

This paper proposes an integrated framework for analyzing the evolution of scientific themes over time by unifying thematic detection and temporal modeling within a single relational structure, improving coherence and robustness.

Contribution

It introduces a novel, structurally consistent approach that models thematic continuity through weighted relational networks, combining lineage reconstruction with cross-sectional analysis.

Findings

Enhanced methodological coherence in longitudinal science mapping

Improved interpretive robustness of thematic evolution analysis

Unified relational framework for detecting and tracking themes over time

Abstract

Strategic diagrams and co-word analysis are widely employed to examine the conceptual structure of scientific domains and their development over time. Yet a structural inconsistency characterises dominant longitudinal implementations: themes are detected through relational clustering in weighted networks, whereas their inter-temporal connections are commonly inferred from set-theoretic overlap among keywords or core documents. This study introduces a structurally integrated framework in which lineage reconstruction is embedded within the same weighted relational architecture that underpins cross-sectional detection. The approach models thematic continuity through graded document affiliation and a lineage-strength measure that combines directional coverage with centrality-weighted structural relevance, thereby conceptualising evolution as the reconfiguration of relational structures rather than simple lexical persistence. By aligning thematic detection and temporal modelling within a unified relational paradigm, the framework enhances the methodological coherence and interpretive robustness of longitudinal science mapping.