Exploring diverse interests of collaborators in smart cities: A topic analysis using LDA and BERT

Area of Science:

• Urban Studies
• Information Science
• Technology Management

Background:

• Urbanization presents complex challenges requiring innovative solutions.
• Smart cities offer a promising framework for addressing urban issues.
• Existing research often lacks a holistic view of diverse smart city stakeholders.

Purpose of the Study:

• To identify key research topics in smart cities by analyzing diverse stakeholder contributions.
• To suggest future research directions for balanced smart city development.
• To address the scarcity of research considering multiple collaborators.

Main Methods:

• Latent Dirichlet Allocation (LDA) for topic modeling.
• Bidirectional Encoder Representations from Transformers (BERT) for text analysis.
• Hype cycle analysis to assess research trends and public expectations.

Main Results:

• Sustainability and digital platforms show high interest across academia, industry, and government.
• Governance, resource management, and green spaces are less emphasized than technology.
• Smart city growth peaked between 2015-2021, with a subsequent slowdown indicating a need for innovation.

Conclusions:

• A significant imbalance exists in research topic distribution among smart city collaborators.
• Future development requires increased focus on underrepresented areas like governance, environment, economy, and healthcare.
• Findings provide actionable insights for researchers, policymakers, and industry to foster comprehensive smart city advancement.