Geographic Space as a Living Structure for Predicting Human Activities Using Big Data

TL;DR

This paper develops a topological, multi-scale representation of geographic space inspired by Christopher Alexander's living structure concept, demonstrating its effectiveness in predicting human activities like tweet locations across scales.

Contribution

It advances the topological representation of geographic space as a multi-scale, living structure, improving prediction of human activities and addressing scale-related issues in traditional geographic models.

Findings

Living structures predict tweet locations effectively across scales.

Topological representation outperforms traditional single-scale models.

Living structure offers a new perspective on geographic space as a multi-scale, organic entity.

Abstract

Inspired by Christopher Alexanders conception of the world - space is not lifeless or neutral but a living structure involving far more small things than large ones a topological representation has been previously developed to characterize the living structure or the wholeness of geographic space. This paper further develops the topological representation and living structure for predicting human activities in geographic space. Based on millions of street nodes of the United Kingdom extracted from OpenStreetMap, we established living structures at different levels of scale in a nested manner. We found that tweet locations at different levels of scale, such as country and city, can be well predicted by the underlying living structure. The high predictability demonstrates that the living structure and the topological representation are efficient and effective for better understanding geographic forms. Based on this major finding, we argue that the topological representation is a truly multi-scale representation, and point out that existing geographic representations are essentially single scale, so they bear many scale problems such as modifiable areal unit problem, the conundrum of length, and the ecological fallacy. We further discuss on why the living structure is an efficient and effective instrument for structuring geospatial big data, and why Alexanders organic worldview constitutes the third view of space. Keywords: Organic worldview, topological representation, tweet locations, natural cities, scaling of geographic space