Scholarlink

<p>This study presents a comprehensive methodology for assessing urban walkability in Batna, Algeria, through the integration of Space Syntax analysis, Geographic Information Systems (GIS), and Machine Learning techniques. The research framework combines subjective perception data collected through the MIT Place Pulse 2.0 model with objective urban characteristics measured at both micro- and macro-scales. The methodology employs convolutional neural networks (CNNs) to quantify subjective walking preferences, while Space Syntax metrics and GIS analysis provide objective measures of urban connectivity and pedestrian infrastructure supporting walkability. A composite walkability index is developed through the weighted aggregation of three distinct indices: the Subjective Walking Preferences Index (SPI), the Micro-Scale Index (MII), and the Macro-Scale Index (MAI). Statistical analysis reveals significant spatial variations in walkability across Batna's neighborhoods (<i>p</i> &lt; 0.001), with strong positive correlations between objective infrastructure quality and subjective perception (r = 0.73, <i>p</i> &lt; 0.001). The integrated approach demonstrates the importance of considering both physical urban characteristics and human perception in walkability assessment, providing evidence-based insights for urban planning and policy development. The proposed framework achieved a prediction accuracy of 84.2% for pedestrian flow patterns, validating its effectiveness for walkability assessment in similar urban contexts.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Urban connectivity and walkability assessment: spatial analysis and predictive modeling—an integrated approach using space syntax, gis, and machine learning for pedestrian flow analysis in batna

  • 期刊论文

  • 发表日期：

    2026年1月7日

查看全文
  • Amina Fantous,
  • Radhia Bennedjai

摘要

This study presents a comprehensive methodology for assessing urban walkability in Batna, Algeria, through the integration of Space Syntax analysis, Geographic Information Systems (GIS), and Machine Learning techniques. The research framework combines subjective perception data collected through the MIT Place Pulse 2.0 model with objective urban characteristics measured at both micro- and macro-scales. The methodology employs convolutional neural networks (CNNs) to quantify subjective walking preferences, while Space Syntax metrics and GIS analysis provide objective measures of urban connectivity and pedestrian infrastructure supporting walkability. A composite walkability index is developed through the weighted aggregation of three distinct indices: the Subjective Walking Preferences Index (SPI), the Micro-Scale Index (MII), and the Macro-Scale Index (MAI). Statistical analysis reveals significant spatial variations in walkability across Batna's neighborhoods (p < 0.001), with strong positive correlations between objective infrastructure quality and subjective perception (r = 0.73, p < 0.001). The integrated approach demonstrates the importance of considering both physical urban characteristics and human perception in walkability assessment, providing evidence-based insights for urban planning and policy development. The proposed framework achieved a prediction accuracy of 84.2% for pedestrian flow patterns, validating its effectiveness for walkability assessment in similar urban contexts.