Demonstrating the effect of exposure to nature on happy facial expressions via Flickr data: Advantages of non-intrusive social network data analyses and geoinformatics methodologies

摘要

Although the role of exposure to nature (ETN) in improving well-being was previously demonstrated, most of the existing research is derived from self-report measures. Conversely, geoinformatics methodologies are seldom used. To address this gap, we examined the prevalence of happy facial expressions (HFE) in natural settings such as water bodies, green vegetation, and undeveloped areas. We applied a novel, spatio-temporal analysis of photos taken in the Boston area and posted on Flickr - a location-based social network - during 2012-2015 (N = 60,013). Photos were analyzed using Microsoft Emotion API to detect facial expressions. ETN, measured either as a composite score, or based on the three aforementioned aspects, was significantly associated with HFEs, even after controlling for temporal patterns. An exploratory visualization of spatial clusters characterized by high HFE proportion was in agreement with the statistical results. This method can be used to explore human-environment interactions more explicitly in a non-intrusive manner.