TLDR: A study introduces a six-dimensional framework to validate and integrate knowledge across AI, urban planning, and social sciences. By analyzing highly cited research from 2014-2024, it identifies dominant perspectives (critical realism, positivism, analytical methods, consequentialism, epistemic values, and social/economic valorization) that enhance research validity, while also highlighting less common but valuable approaches. The framework aims to guide researchers in fostering credible and socially impactful transdisciplinary collaborations for addressing complex urban challenges.
Addressing the complex challenges facing our cities today, from climate change to social inequality, increasingly requires a collaborative approach that goes beyond traditional academic boundaries. This is especially true when combining the powerful capabilities of Artificial Intelligence (AI) with the nuanced understanding of urban planning and social sciences. However, effectively bringing together these distinct fields, with their different ways of thinking and working, can be quite difficult.
A recent research paper, titled Intersectoral Knowledge in AI and Urban Studies: A Framework for Transdisciplinary Research by Rashid Mushkani, proposes a groundbreaking six-dimensional framework designed to help researchers validate and integrate knowledge across AI and city studies. This framework is built upon an extensive analysis of highly cited research from 2014 to 2024, offering a guide for fostering more credible and socially impactful transdisciplinary collaborations.
Understanding the Framework’s Dimensions
The framework classifies research approaches across six key dimensions, each representing a fundamental aspect of how knowledge is created and understood:
- Ontological: This dimension explores the nature of reality being studied. For instance, does reality exist independently of our perception, or is it mentally constructed?
- Epistemological: This focuses on how knowledge is validated. Is it through observation and experiment, or through reason and interpretation?
- Methodological: This refers to the research methods used, such as quantitative (numbers and statistics), qualitative (meanings and experiences), or a mix of both.
- Teleological: This considers the purpose and goals of the research. Is the focus on the outcomes of actions, or on duties and principles?
- Axiological: This addresses the values and principles guiding the research, such as truth and objectivity (epistemic), moral rightness (ethical), or beauty and creativity (aesthetic).
- Valorization: This dimension looks at how the broader value of research outputs is recognized and enhanced, whether for economic benefits, societal well-being, or cultural preservation.
How the Study Was Conducted
To develop this framework, the researcher conducted a comprehensive literature search using the Scopus database, focusing on publications from 2014 to 2024. The study identified 2,000 of the most cited papers across four domains: AI in computer science, city in computer science, city in social science, and AI in social science. An advanced AI model, GPT-4o, was then used to classify each paper according to the six dimensions, ensuring a systematic and consistent analysis.
Key Findings: Dominant and Emerging Perspectives
The analysis revealed several dominant perspectives consistently adopted across the sampled literature. These include:
- Critical Realism (Ontological): Acknowledging an objective reality while recognizing that social and historical contexts influence how we understand it.
- Positivism (Epistemological): Viewing scientific, mathematically treated knowledge as the primary legitimate form.
- Analytical Methods (Methodological): Relying on logical or conceptual analysis.
- Consequentialism (Teleological): Evaluating actions based on their outcomes.
- Epistemic Values (Axiological): Prioritizing truth and objectivity in research.
- Social and Economic Valorization (Valorization): Focusing on improving societal well-being or translating findings into market/industry benefits.
These dominant viewpoints align with existing norms in both computer science and social science, suggesting that adopting them can enhance the acceptance of intersectoral research. However, the study also highlighted less common perspectives, such as idealism (ontological), rationalism (epistemological), mixed methods (methodological), pragmatism (teleological), aesthetic values (axiological), and cultural valorization. While less established, these approaches offer the potential to bring innovative insights, particularly by incorporating humanistic and contextual dimensions often overlooked in AI-driven urban research.
Also Read:
- EMPATHIA: A New AI Framework for Dignified Refugee Integration
- Understanding AI’s Biggest Threats: A New Framework for Risk Management
Implications for Future Research
The framework provides practical guidance for researchers and stakeholders. It encourages a critical self-assessment of one’s own perspectives (reflexive alignment) to better engage with other disciplines. Active stakeholder engagement, involving policymakers, practitioners, and community members from the outset, is also crucial to ensure research questions and outcomes are relevant and actionable. Furthermore, implementing continuous feedback mechanisms and transparent evaluation criteria can help overcome biases against intersectoral projects and foster recognition of diverse contributions.
As AI becomes increasingly integrated into urban policy and planning, collaborative efforts that embrace diverse ways of knowing and valuing are essential. This framework serves as a valuable tool for building transdisciplinary teams that can enhance both the rigor and social accountability of research, ultimately contributing to more inclusive and sustainable urban futures.
