According to the 2011 Census, 2.21% of India’s population comprises persons with disabilities, with 30.5% residing in urban areas. However, urban infrastructure remains largely inaccessible, as 85.5% and 61.7% of urban persons with disabilities reported difficulties in accessing public transportation and public buildings, respectively. This research explores the use of emerging technologies such as GIS and Machine Learning to assess the universal accessibility of built environments in Indian cities, with Madurai (Tamil Nadu) as the case study. GIS was employed to evaluate the accessibility of tourist sites from the perspective of visitors with mobility impairments. Accessibility maps were created using access audit data, based on checklists developed by the authors in 2020 for city entry points and public spaces. GIS mapping demonstrated several advantages, including replicability across cities and the facilitation of data sharing among agencies for informed decision-making on retrofitting. A major challenge encountered was the need for frequent data updates, as physical audits are both time-consuming and costly. To address this, a Machine Learning approach using sentiment analysis was adopted. Google reviews of tourist sites in Madurai were analyzed using the VADER tool to extract accessibility-related sentiments. Reviews were classified as positive, negative, or neutral based on compound sentiment scores and were aggregated to determine site-level accessibility perceptions. The sentiment analysis results aligned with the GIS findings, validating the effectiveness of Machine Learning in capturing real-time accessibility information. This integrated approach offers a scalable solution for assessing and enhancing universal accessibility in Indian cities.