Abstract

Urban environments present significant ongoing barriers to physical activity and public health, including motivational challenges, environmental constraints, uniformity of existing fitness apps, and limited opportunities for social engagement. This project proposes an AI-powered Walkability Assistant designed to promote healthier, more active lifestyles through personalized, adaptive, and community-driven solutions.

This study advances past efforts by introducing a novel wellbeing framework that encompasses four dimensions: physical, social, environmental, and psychological wellbeing. To identify user-driven strengths and areas for improvement, we conducted sentiment analysis (positive/negative) and Latent Dirichlet Allocation (LDA) topic modeling on large-scale user-generated data from walking and fitness app reviews, leveraging a custom wellbeing taxonomy to categorize users’ perceptions and preferences. This layered approach revealed distinct needs and improvement areas across the four wellbeing dimensions, supporting more targeted recommendations and adaptive interventions. Ongoing open-ended survey results, informed by our framework, continue to provide rich qualitative perspectives that complement large-scale findings.

Building on these findings, the proposed prototype incorporates personalized route recommendations, real-time adaptation to environmental conditions, and gamified incentives to sustain user engagement. The Walkability Assistant is designed to decrease sedentary behavior, enhance holistic wellbeing, and foster inclusive urban communities through AI-driven, dynamically adaptive features. By synthesizing user-centered design and scalable data science, this project makes a substantive contribution to public health and sustainable urban communities.

Keywords

Artificial Intelligence, Walkability, Public Health, Urban Mobility, User-Centered Design

Date of this Version

11-18-2025

Download

Share

COinS