Smart City Data Analysis via Visualization of Correlated Attribute Patterns

TL;DR

This paper presents a visualization system for analyzing correlated attribute patterns in smart city sensor data, enabling intuitive understanding of urban conditions through spatial and temporal correlation visualization.

Contribution

The paper introduces a novel visualization system for correlated attribute pattern mining in smart city data, integrating spatial and temporal analysis for better urban insights.

Findings

System supports interactive analysis of smart city data.

Effective visualization of sensor correlations on maps and over time.

Demonstrated on datasets from China and Spain.

Abstract

Urban conditions are monitored by a wide variety of sensors that measure several attributes, such as temperature and traffic volume. The correlations of sensors help to analyze and understand the urban conditions accurately. The correlated attribute pattern (CAP) mining discovers correlations among multiple attributes from the sets of sensors spatially close to each other and temporally correlated in their measurements. In this paper, we develop a visualization system for CAP mining and demonstrate analysis of smart city data. Our visualization system supports an intuitive understanding of mining results via sensor locations on maps and temporal changes of their measurements. In our demonstration scenarios, we provide four smart city datasets collected from China and Santander, Spain. We demonstrate that our system helps interactive analysis of smart city data.