Social Media Analytics

by

Bo Zhao

• LAST REVIEWED: 26 October 2023
• LAST MODIFIED: 26 October 2023
• DOI: 10.1093/obo/9780199874002-0275

Introduction

Social media analytics is the process of collecting, analyzing, and visualizing data from platforms like Twitter, Instagram, Weibo, and TikTok to gain insights into how people behave in the real world or virtual spaces. The data is a collection of digital records people create while interacting with social media platforms. Each record often contains textual or multimedia descriptions of the user’s input, a unique identifier, a geotag indicating the interaction’s location, and a timestamp indicating when the interaction occurs. In practice, social media data is collected in large quantities via front-end scraping or an Application Programming Interface (API). Once collected, the data can be examined using content analysis, spatial-temporal analysis, and geovisual analysis, although these analyses are often mutually inclusive. Content Analysis involves examining the metadata, text, images, or videos in social media posts to extract insights. This can be done using keyword matching, machine learning, or topic modeling. Spatial-Temporal Analysis examines the location and timing of social media posts, and can be used to identify event patterns, predict future occurrences, find geographical clusters, assess spatial-temporal correlations, examine patterns in social networks, and determine spatial-temporal variation in hotspots of activity. Geovisual Analytics involves creating maps or other visual representations of social media data to help understand patterns and trends. These analyses can be combined to provide a comprehensive understanding of the phenomena depicted in the data. However, it’s important to note that using social media data raises social and ethical concerns when such data is presented, curated, or commodified. Some identified issues include data bias, privacy leaks, and location spoofing.

General Overview

Social media has been influencing different aspects of human society, both by connecting us on the ground and by linking us to the virtual world of cyberspace. Goodchild 2007 argues that this capacity of social media enables each user to behave like a sensor, collecting and sharing data. With the convergence of the mobile Internet, the Global Positioning System (GPS), and high-performance computing, it becomes possible to analyze a large amount of data from social media such as Twitter, Instagram, Weibo, and TikTok. Through social media analytics, data scientists and industry experts can learn more about how people act on the ground and connect in the real world and online spaces. Stefanidis, et al. 2013 offered an early perspective on the potential for collecting and utilizing geographical data from social media. Several systematic literature reviews—conducted in Batrinca and Treleaven 2015; Stock 2018; and Ghani, et al. 2019—summarize how social media data has been collected, analyzed, and visualized. Because social media data contains textual or multimedia content and spatial and temporal metadata, there are different ways to analyze social media depending on the data being examined. So three types of analysis have been identified, although they are often mutually inclusive. They are content analysis, spatial-temporal analysis, and geovisual analysis. Combining these analyses can develop a comprehensive understanding of the phenomena the social media data depict. For example, Zou, et al. 2018 traced and mapped disease outbreaks using Twitter data; Kay, et al. 2015 explored public perceptions of air pollution using a Chinese social media; Fekete and Haffner 2019 examined the trends in the geographic discipline by analyzing the Twitter data with the #AAG2018 hashtag; and Tsao, et al. 2021 provided a review of how social media data were used to analyze the COVID-19 spreading tendencies. It should be emphasized that this entry Social Media Analytics has focused on analyzing social media data from various platforms, particularly those that are popular in certain countries such as Weibo in China. While some representative papers on Weibo analysis are reviewed in this article, it is important to acknowledge that a significant portion of the literature on this topic has not been published in Western journals. Moreover, while social media analysis can be prominent in the study of human activity and geographical events, both spatially and temporally, it also unavoidably produces new challenges for different disciplines, like GIScience, as argued in Sui and Goodchild 2011. Also, it is urgent to consider the ethical issues it may raise, as implied by the study Gerbaudo 2018.

• Batrinca, B., and P. C. Treleaven. “Social Media Analytics: A Survey of Techniques, Tools, and Platforms.” AI & Society 30.1 (2015): 89–116.

  DOI: 10.1007/s00146-014-0549-4

  This paper reviews software tools for analyzing social media, including scraping, storage, data cleaning, and sentiment analysis. It also provides a methodology and critique of these tools and discusses the rapidly evolving commercial landscape of social media data services. The paper specifically focuses on Twitter in particular for sentiment analysis and provides an overview, including code fragments, for scientists interested in using social media scraping and analytics.

• Fekete, E., and M. Haffner. “Twitter and Academic Geography through the Lens of #AAG2018.” The Professor Geographer 71.4 (2019): 751–761.

  DOI: 10.1080/00330124.2019.1622428

  The authors dissect the collected Twitter data, using tools like word clouds and sentiment analysis, to uncover trends and sentiments associated with the AAG Annual Meeting. The study not only offers valuable insights into the implementation of social media analytics, but also suggests that Twitter data could serve as a barometer for evolving trends in the geographic discipline.

• Gerbaudo, P. “Social Media and Populism: An Elective Affinity?” Media, Culture & Society 40.5 (2018): 745–753.

  DOI: 10.1177/0163443718772192

  The author argues that the popularity of social media among populist politicians is due to its ability to facilitate mass networking and politics. The article examines the role of social media in populist movements as a means for individuals to express themselves and for like-minded individuals to gather and form online crowds.

• Ghani, N. A., S. Hamid, I. A. T. Hashem, and E. Ahmed. “Social Media Big Data Analytics: A Survey.” Computers in Human Behavior 101 (2019): 417–428.

  DOI: 10.1016/j.chb.2018.08.039

  This study reviews recent works on social media analytics and compares big data analytics techniques and their quality attributes. The study also discusses the applications of social media big data analytics, highlighting state-of-the-art techniques and methods, as well as open research challenges in this field.

• Goodchild, M. F. “Citizens as Sensors: The World of Volunteered Geography.” GeoJournal 69.4 (2007): 211–221.

  DOI: 10.1007/s10708-007-9111-y

  This paper reviews the rise of social media and how the data it contributes can be used for various purposes. It also discusses the motivations, accuracy, potential privacy threats, and how social media may augment more conventional sources. Social media is considered a data source for understanding the Earth’s surface and local activities that may not be reported by mainstream media.

• Kay, S., B. Zhao, and D. Z. Sui. “Can Social Media Clear the Air? A Case Study of the Air Pollution Problem in Chinese Cities.” The Professional Geographer 67.3 (2015): 351–363.

  DOI: 10.1080/00330124.2014.970838

  A sample of microblog posts about air pollution in China from 2012–2013, along with contextual information from various sources, is analyzed to study how the government, companies, NGOs, and individuals approach the Chinese social media landscape. The research finding implies that while microblogs can empower citizens to advocate for environmental causes, social media are also used by the government for social monitoring and control and by companies for profit.

• Stefanidis, A., A. Crooks, and J. Radzikowski. “Harvesting Ambient Geospatial Information from Social Media Feeds.” GeoJournal 78.2 (2013): 319–338.

  DOI: 10.1007/s10708-011-9438-2

  Social media data often contains geographic footprints. This paper proposes a framework for harvesting and analyzing this “ambient” geospatial information from social media to support situational awareness of human activities. This marks the second step in the evolution of geospatial data availability following the emergence of volunteered geographical information.

• Stock, K. “Mining Location from Social Media: A Systematic Review.” Computers, Environment and Urban Systems 71 (2018): 209–240.

  DOI: 10.1016/j.compenvurbsys.2018.05.007

  This systematic review finds that while coordinates are often used in social media posts, they have limited coverage and do not always reflect the location referred to in the post. Place name extraction provides the most accurate results, but all methods have limitations. This review also analyzes different social media platforms and identifies potential future areas.

• Sui, D. Z., and M. F. Goodchild. “The Convergence of GIS and Social Media: Challenges for GIScience.” International Journal of Geographical Information Science 25.11 (2011): 1737–1748.

  DOI: 10.1080/13658816.2011.604636

  This paper discusses the new challenges for GIScience due to the convergence of GIS and social media. It notes that remarkable conceptual and technological advances in GIS have occurred over the past decade and highlights the need to consider GIScience in relation to media studies theories such as Marshall McLuhan’s laws of the media.

• Tsao, S.-F., H. Chen, T. Tisseverasinghe, Y. Yang, L. Li, and Z. A. Butt. “What Social Media Told Us in the Time of COVID-19: A Scoping Review.” The Lancet Digital Health 3.3 (2021): e175–e194.

  DOI: 10.1016/S2589-7500(20)30315-0

  This paper finds that social media plays a significant role in surveying public attitudes, identifying infodemics, assessing mental health, detecting or predicting COVID-19 cases, analyzing government responses, and evaluating the quality of health information in prevention education videos. Moreover, this paper also highlights the possibility of combating misinformation related to COVID-19 using social media data.

• Zou, L., N. S. Lam, H. Cai, and Y. Qiang. “Mining Twitter Data for Improved Understanding of Disaster Resilience.” Annals of the American Association of Geographers 108.5 (2018): 1422–1441.

  DOI: 10.1080/24694452.2017.1421897

  This study analyzes Twitter data from Hurricane Sandy in the US Northeast to identify common indexes that can be used for emergency management and resilience analysis. Moreover, this study also examines geographical disparities in disaster-related Twitter data and tests whether Twitter data can improve postdisaster damage estimation. The results indicate that social media data indeed could help in that regard.