overview: Researchers were able to use Twitter data to identify specific events and locations within the city associated with different emotions. Stations and public transport locations tended to be less joyful and more disgusting, while hotels and restaurants tended to be more joyful expressions.
sauce: PLOS
An analysis of nearly two million tweets made by people in London and San Francisco explored specific event and place types associated with different emotions.
Panote Siriaraya, of Kyoto Institute of Technology, Japan, et al. pro swan.
An increasing number of studies are examining social media posts and location data to explore human behavior and emotions. For example, when comparing happiness across regions.
However, much of that work has been confined to larger geographic scales, focusing on only one emotion at a time or focusing on general assessments of positive and negative emotions. increase.
Siriaraya and colleagues now show how to use tweets and information about specific buildings, businesses, and other places of interest from the public platform Open Street Map to explore human emotional expressions at a more granular level. I’m here.
Using a computational tool called a neural network, they analyzed about 2 million tweets posted by more than 200,000 people in London and San Francisco to find out when and where people were angry, hopeful, disgusted, feared, joyful and sad. , surprised, expressed confidence, or identified.
Analysis showed that different location types were associated with different emotional expressions. For example, in both cities, tweets made at train stations, bridges, and other forms of transport tended to express disgust rather than joy. Tweets from hotels and restaurants showed higher levels of joy. Furthermore, proximity to a particular site, rather than just being within the site, was associated with differences in the emotions expressed.
Certain events appeared associated with certain higher-level emotions. For example, users in San Francisco indicated the highest levels of anger, disgust, and sadness on her 2017 Women’s March Day, while users in London indicated the highest levels of fear and grief during her two local terrorist attacks. showed sadness. New Year’s Eve coincided with a high level of joy in both cities.
The researchers caution against overgeneralizing their results. For example, this study included only English-language tweets. Nonetheless, it could help pave the way for additional fine-grained research to inform fields such as urban planning and tourism.
The authors conclude, “Our study highlights how publicly available data sources can be used to portray fine-grained emotional signatures at detailed spatial and temporal levels across cities. .”
About this social media and sentiment research news
author: press office
sauce: PLOS
contact: Press Office – PLOS
image: image is public domain
See also
Original research: open access.
“A Citywide Survey of Granular Human Emotions with Social Media Analysis” by Panote Sirialaya et al. pro swan
overview
A city-wide survey of fine-grained human emotions with social media analytics
The proliferation of social media and open web data has provided researchers with a unique opportunity to better understand human behavior on many levels. In this white paper,
Open Street Map and Twitter can be analyzed and used to detail human emotions at a city-wide level in two cities, San Francisco and London. A fine-grained sentiment neural network classifier was developed, tested, and used to detect sentiment from tweets in two cities. Detected emotions were matched against key locations extracted from Open Street Map.
Through analysis of the obtained datasets, we highlight the influence of different days, locations and POI neighborhoods on the representation of human emotions in the city.
