Advances in AI offer new ways to integrate citizen perspectives into urban decision-making, addressing the persistent challenge of scarce or missing participation data in many cities. This paper presents an NLP–AHP framework that operationalizes unstructured text into structured decision criteria for smart-city planning. Using online park reviews from different contexts as a case study, Natural Language Processing (NLP) is used to extract features from externally sourced data; then mapped to the Analytic Hierarchy Process (AHP) as normalized pairwise judgments to produce criteria weights. The approach demonstrates how this computationally intelligent framework ensures perfect consistency in the decision matrix, providing an evidence-based proxy for the “citizen’s voice” when conventional participation is infeasible. Applications extend beyond parks to broader urban domains such as housing, mobility, and public space design in resource-limited contexts. This study contributes by formalizing a bridge between text-mined insights and structured multi-criteria weights, expanding the participatory planning toolkit with an AI-driven layer. While not a substitute for genuine local engagement, the framework offers a scalable and adaptable proxy until richer participation data can be collected, enabling more inclusive and evidence-based city planning.