Human Emotions at Different Places: A Ranking of Happiest Tourist Attractions around the World Based on Facial Expressions and Spatial Clustering Analysis

Authors: Yuhao Kang*, University of Wisconsin Madison, Song Gao, University of Wisconsin Madison
Topics: Geographic Information Science and Systems
Keywords: affective computing, spatial clustering, user generated data, place-based GIS
Session Type: Paper
Day: 4/4/2019
Start / End Time: 5:00 PM / 6:40 PM
Room: 8228, Park Tower Suites, Marriott, Lobby Level
Presentation File: No File Uploaded


Understanding people’s perception to environment is a key aspect of human behavior analysis and a core issue in human geography. Scholars from social sciences usually use questionnaire to investigate the emotion of people in different environmental contexts, which costs a lot of human resources and lacks timeliness. With the emergence of spatio-temporal big data and state-of-the-art artificial intelligence (AI) technologies, it is now possible to evaluate various human emotions at different places from a statistic perspective by applying affective computing to huge amount of user generated data in social media. And with the ubiquitous use of positioning devices, those generated large-scale georeferenced data could help locate where people express their emotions.

In this study, we proposed a novel framework to measure human emotions and to explore potential causing factors to the degree of happiness at different places. Different from previous researches which mainly utilized text-based sentiment perception systems via NLP technologies, we investigated image-based emotion extraction methods across all kinds of human facial expressions. Compared with text-based approaches that face challenges like linguistic ambiguity and multi-cultural translation, the facial expression of representing emotions is said to be universal across countries and different periods (e.g., the ancestor of human like chimpanzees have similar emotions with modern people). Therefore, facial expression-based emotion detection may be much more suitable for global-scale issues. Moreover, we further analyze the relationship between different kinds of geographical contexts and the degree of happiness extracted from human facial expressions at different places.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login