Cite this as: Tenzer, M. 2022 Tweets in the Peak: Twitter Analysis - the impact of Covid-19 on cultural landscapes, Internet Archaeology 59. https://doi.org/10.11141/ia.59.6
The Covid-19 pandemic had an unprecedented impact on society, with restrictions on socialising and movement during the three lockdown periods between March 2020 and March 2021 (Baker et al. 2021; Institute for Government Analysis 2021). Easily accessible locations offering the typical qualities of tourist destinations moved into the focus of day visitors in periods when restriction eased. The Peak District National Park (PDNP), a cultural landscape comprising historical places, natural beauty spots, and 'chocolate box' villages, offered a way of satisfying the urge to escape to the countryside. The impact was also felt in the heritage sector, with a noticeable change in visitor behaviour and the relationship between park residents and day tourists (Jones and McGinlay 2020; Sofaer et al. 2021).
In order to understand societal change, social media research gives a unique insight into the sentiments, actions, and controversies associated with tourism, Covid-19, and nature conservation. In particular, the open and public nature of Twitter data offers itself for the analysis of large datasets based on specific search queries at specific time periods.
For this research, tweets from the PDNP for three weekends in 2019 to 2021 with different restriction levels were collected. Using R and Python, automated processes allow the time-efficient analysis of qualitative information. This project has extended the standard procedures of social media analysis, such as keyword search and sentiment analysis by an emoji analysis and location entity recognition, focusing specifically on cultural and natural heritage. Using Twitter data in a time-efficient process and creating visually appealing outputs may foster an appreciation of the park's resources and positively influence the behaviour of visitors and residents. Going forward, improving the relationship between people and places will provide background for the management of cultural landscapes and help tackle environmental issues, such as peat erosion resulting from a large influx of walkers, address the climate change emergency, and help ease the controversial relationship between a living and working landscape and tourism.
Corresponding author: Martina Tenzer
University of York
Figure 1: Study area: Peak District National Park (map created by M. Tenzer, basemap © OpenStreetMap contributors)
Figure 2: Network graph visualisation of word associations based on frequently used words, such as 'walk', 'see', 'view', 'honeypot', 'landscape', 'place', as well as words of special interest regarding heritage management, such as 'heritage', 'history', and 'monument'. The colours represent the three years. The visualisation represents statistical associations of the words
Figure 3: Word cloud of most frequently used words across the study period. Disclaimer: The words and their position are not geospatially located within the Peak District National Park, but rather randomly located within the boundary to visualise the summary and frequency of the most used words in tweets related to the area within these boundaries
Figure 4: Sentiment analysis with VADER sentiment algorithm based solely on emojis. The emoji score is normalised for the sentiment categories of the respective years. Note: the category 'Neutral' has been excluded from the visualisation, as this would skew the plot, as this category combined all the tweets with no emojis, neutral emojis and unknown emojis
Figure 5: Cloud of most frequently used emojis across the study period. Disclaimer: the emojis and their position are not geospatially located within the Peak District National Park, but rather randomly located within the boundary to visualise the summary and frequency of the most used emojis in tweets related to the area within these boundaries
Figure 6: Sentiment analysis with VADER sentiment algorithm. The score is normalised for the sentiment categories of the respective years
Figure 7: Confusion matrix comparing manual and automated place recognition for the year 2020. Bottom left and top right quadrants show the number of True-Negative and True-Positive results of the recognition algorithm, respectively, top left and bottom right show False-Positive and False-Negative matches, respectively
Figure 8: Confusion matrix comparing manual and automated place recognition for the year 2019. Bottom left and top right quadrants show the number of True-Negative and True-Positive results of the recognition algorithm, respectively, top left and bottom right show False-Positive and False-Negative matches, respectively
Figure 9: Confusion matrix comparing manual and automated place recognition for the year 2021. Bottom left and top right quadrants show the number of True-Negative and True-Positive results of the recognition algorithm, respectively, top left and bottom right show False-Positive and False-Negative matches, respectively
Figure 10: Building a corpus of locations in GIS gradually in steps, adding levels of information from various sources, such as rivers, place names, points of interest from Ordnance Survey data, or historic information from Historic England data (Image: M. Tenzer)
Figure 11: Confusion matrix, comparing manual and automated place recognition. Bottom left and top right quadrants show the number of True-Negative and True-Positive results of the recognition algorithm, respectively. Top left and bottom right show False-Positive and False-Negative matches, respectively
Figure 12: QGIS2Web OpenLayers map showing point data and heatmaps of locations mentioned in tweet texts of the Spring bank holiday weekends 2019 to 2021. Base maps used in this project provide the background for orientation and navigation of the map. For map use, please zoom in for details, and switch layers on and off in the top right corner menu
Figure 13: Heatmap visualisation of the years 2019 to 2021 showing the shift of locations mentioned in tweets visualised in QGIS (map created by M. Tenzer, basemap map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL)
Figure 14: Another option for visualising hot spot locations mentioned in tweets for the years 2019-2021 in one 2D map: point size based on weight (frequency) visualised in QGIS (map created by M. Tenzer, basemap map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL)
Table 1: Top 40 of the most frequently used hashtags across the study period
Table 2: Top 30 of the most frequently used words across the study period
Table 3: Top 20 of the most frequently used emojis across the study period
Table 4: User location of tweets about the Peak District National Park with the 25 most frequently used location descriptions
Baker, C., Kirk-Wade, E., Brown, J. and Barber, S. 2021 'Coronavirus: a history of English lockdown laws', House of Commons Library. https://commonslibrary.parliament.uk/research-briefings/cbp-9068/
Barai, M.K. 2021 'Sentiment analysis with Textblob and Vader in Python', Analytics Vidhya. https://www.analyticsvidhya.com/blog/2021/10/sentiment-analysis-with-textblob-and-vader/
Barnatt, J. and Penny, R. 2004 The Distribution of Lead Mining Surface Remains in the Peak District, Bakewell: Peak District National Park Authority.
Barrie, C. 2022 'academictwitteR'. https://github.com/cjbarrie/academictwitteR
Barrie, C. and Chun-ting Ho, J. 2021 'academictwitteR: an R package to access the Twitter Academic Research Product Track V2 API Endpoint', Journal of Open Source Software 6(62). https://doi.org/10.21105/joss.03272
Benesch, S. 2021 'Nobody can see into Facebook', The Atlantic. https://www.theatlantic.com/ideas/archive/2021/10/facebook-oversight-data-independent-research/620557/
Bertrand, K.Z., Bialik, M., Virdee, K., Gros, A. and Bar-Yam, Y. 2013 'Sentiment in New York City: a high resolution spatial and temporal view', arXiv:1308.5010 [Physics], August. http://arxiv.org/abs/1308.5010
Bianchini, R. 2021 'Museums worldwide react to COVID lockdown by offering virtual tours | Inexhibit'. https://www.inexhibit.com/marker/museums-worldwide-react-to-covid-lockdown-by-offering-virtual-visits/
Cao, N. and Cui, W. 2016 Introduction to Text Visualization, Paris: Atlantis Press. https://doi.org/10.2991/978-94-6239-186-4
Drus, Z. and Khalid, H. 2019 'Sentiment analysis in social media and its application: systematic literature review', Procedia Computer Science 161, 707–14. https://doi.org/10.1016/j.procs.2019.11.174
Edelson, L. and McCoy, D. 2021 'Facebook is obstructing our work on disinformation. other researchers could be next', The Guardian, 14 August 2021. https://www.theguardian.com/technology/2021/aug/14/facebook-research-disinformation-politics
English Heritage 2000 Power of Place: The Future of the Historic Environment, London: English Heritage.
Farrow, J. 2021 'May 2021 - cold and wet - saved only by the Bank Holiday Weekend', NetweatherTV, 1 June 2021. https://www.netweather.tv/weather-forecasts/news/10895-may-2021---cold-and-wet---saved-only-by-the-bank-holiday-weekend
Gajadhar, J. and Green, J. 2005 'The importance of nonverbal elements in online chat', Educause Quarterly 28(4), 63-64. https://www.learntechlib.org/p/103672/
Gibney, E. 2019 'Privacy hurdles thwart Facebook democracy research', Nature 574(7777), 158–59. https://doi.org/10.1038/d41586-019-02966-x
Gutowski, P. and Kłos-Adamkiewicz, Z. 2020 'Development of e-Service virtual museum tours in Poland during the SARS-CoV-2 pandemic', Procedia Computer Science, Knowledge-Based and Intelligent Information & Engineering Systems: Proceedings of the 24th International Conference KES2020, 176 (January), 2375–83. https://doi.org/10.1016/j.procs.2020.09.303
Hegelich, S. 2020 'Facebook needs to share more with researchers', Nature 579(7800), 473–73. https://doi.org/10.1038/d41586-020-00828-5
Hern, A. 2015 'Don't know the difference between Emoji and Emoticons? Let me explain', The Guardian, 6 February 2015. https://www.theguardian.com/technology/2015/feb/06/difference-between-emoji-and-emoticons-explained
Historic England 2022 'Download listing data - GIS shapefiles'. http://historicengland.org.uk/listing/the-list/data-downloads/
Hutto, C.J. 2022 'Cjhutto/vaderSentiment'. https://github.com/cjhutto/vaderSentiment
Hutto, C. and Gilbert, E. 2014 'VADER: a parsimonious rule-based model for sentiment analysis of social media text', Proceedings of the International AAAI Conference on Web and Social Media 8(1), 216–25. https://ojs.aaai.org/index.php/ICWSM/article/view/14550
Iglesias, C. and Moreno, A. (eds) 2020 Sentiment Analysis for Social Media, MDPI. https://www.mdpi.com/books/pdfview/book/2154
Institute for Government Analysis 2021 'Timeline-Lockdown'. https://www.instituteforgovernment.org.uk/sites/default/files/timeline-lockdown-web.pdf
Jones, N. and McGinlay, J. 2020 The Impact of COVID-19 Restrictions on Local Communities of Peak District National Park and Management Options During the Pandemic, Cambridge: University of Cambridge.
Linebaugh, K. and Knutson, R. 2021 'The clash between Facebook and independent researchers - The Journal. - WSJ Podcasts', 9 August 2021. [Last accessed: 11 February 2022]. https://www.wsj.com/podcasts/the-journal/the-clash-between-facebook-and-independent-researchers/8fda97fc-203d-4632-bc86-f81c0cbe5faf
Low, S.M. 2002 'Anthropological-ethnographic methods for the assessment of cultural values in heritage conservation' in Marta de la Torre (ed) Assessing the Values of Cultural Heritage: Research Report, Los Angeles: The Getty Conservation Institute. 31–49.
Lynch, K. 1960 The Image of the City, Cambridge: THe MIT Press & Harvard University Press.
Madgin, R. 2021 'Emoji as method' in R. Madgin and J. Lesh (eds) People-Centred Methodologies for Heritage Conservation, London: Routledge. 80–94. https://doi.org/10.4324/9780429345807-6
Madgin, R. and Lesh, J. 2021 People-Centred Methodologies for Heritage Conservation, London: Routledge. https://doi.org/10.4324/9780429345807
Mahmud, J., Nichols, J. and Drews, C. 2014 'Home location identification of Twitter users', ACM Transactions on Intelligent Systems and Technology 5(3), 1–21. https://doi.org/10.1145/2528548
Malde, R. 2020 'A short introduction to VADER', Medium. https://towardsdatascience.com/an-short-introduction-to-vader-3f3860208d53
Messina, C. n.d. 'Chris Messina', Chris Messina. https://chrismessina.me [Last accessed: 6 July 2022].
Meta 2021 'The Facebook Company is now Meta', Meta. https://about.fb.com/news/2021/10/facebook-company-is-now-meta/
Meta 2022 'Academic resources - Meta Research | Meta Research'. https://research.facebook.com/data/
MetOffice 2019 'Weather this Bank Holiday Weekend?' Met Office, 23 May 2019. https://www.metoffice.gov.uk/about-us/press-office/news/weather-and-climate/2019/late-may-bank-holiday-2019
MetOffice 2020 'May 2020 becomes the sunniest calendar month on record', Met Office, 1 June 2020. https://www.metoffice.gov.uk/about-us/press-office/news/weather-and-climate/2020/2020-spring-and-may-stats
Milgram, S. and Jodelet, D. 1992 'Psychological Maps of Paris' in S. Milgram (ed) The individual in a social world. Essays and Experiments, 2nd edition, McGraw-Hill. 88–113.
National Trust 2017 'Places That Make Us Research Report', Places That Make Us, Swindon: National Trust. https://nt.global.ssl.fastly.net/documents/places-that-make-us-research-report.pdf
Neri, F., Aliprandi, C., Capeci, F., Cuadros, M. and By, T. 2012 'Sentiment analysis on social media'. https://doi.org/10.1109/ASONAM.2012.164
Novak, P.K., Smailović, J., Sluban, B. and Mozetič, I. 2015 'Sentiment of emojis', PLOS ONE 10(12), e0144296. https://doi.org/10.1371/journal.pone.0144296
Ordnance Survey n.d. 'Free OS OpenData map downloads | Free vector & raster map data | OS data hub', OS OpenData Downloads. https://osdatahub.os.uk/downloads/open
Oxford Languages 2022a 'Oxford word of the year 2015 | Oxford languages'. https://languages.oup.com/word-of-the-year/2015/
Oxford Languages 2022b 'Oxford word of the year 2020 | Oxford languages'. https://languages.oup.com/word-of-the-year/2020/
Panko, B. 2017 'A decade ago, the hashtag reshaped the Internet', Smithsonian Magazine. https://www.smithsonianmag.com/smart-news/decade-ago-hashtag-reshaped-internet-180964605/
PDNPA n.d. History of Our National Park: Peak District National Park. http://www.peakdistrict.gov.uk/learning-about/about-the-national-park/our-history
Pulido, C., Redondo-Sama, G., Sordé Martí, T. Flecha, R. 2018 'Social impact in social media: a new method to evaluate the social impact of research', PLOS ONE 13 (August), e0203117. https://doi.org/10.1371/journal.pone.0203117
RDocumentation n.d. 'findAssocs Function - RDocumentation'. https://www.rdocumentation.org/packages/tm/versions/0.7-8/topics/findAssocs
Ritter, A. and Clark, S. 2011 'Named Entity Recognition in Tweets: An Experimental Study' in Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing. Presented at the Conference on Empirical Methods in Natural Language Processing, Association for Computational Linguistics, Edinburgh, Scotland, UK. 1524–1534. https://aclanthology.org/D11-1141.pdf
Samaroudi, M., Rodriguez Echavarria, K. and Perry, L. 2020 'Heritage in lockdown: digital provision of memory institutions in the UK and US of America during the COVID-19 pandemic', Museum Management and Curatorship 35(4), 337–61. https://doi.org/10.1080/09647775.2020.1810483
Samuels, A. and Mcgonical. J. 2020 'Sentiment analysis on social media content',' arXiv:2007.02144 [Cs], July. http://arxiv.org/abs/2007.02144
Social Media Research Group 2016 GSR Social Media Research Guidance - Using social media for social research. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/524750/GSR_Social_Media_Research_Guidance_-_Using_social_media_for_social_research.pdf
Sofaer, J., Davenport, B., Sørensen, M.L.S., Gallou, E. and Uzzell, D. 2021 'Heritage sites, value and wellbeing: learning from the COVID-19 pandemic in England', International Journal of Heritage Studies 27(11), 1117–32. https://doi.org/10.1080/13527258.2021.1955729
Stamp, J. 2013 'Who really invented the smiley face?', Smithsonian Magazine. https://www.smithsonianmag.com/arts-culture/who-really-invented-the-smiley-face-2058483/
Taplin, D.H., Scheld, S. and Low, S.M. 2002 'Rapid ethnographic assessment in urban parks: a case study of Independence National Historical Park', Human Organization 61(1), 80–93. https://doi.org/10.17730/humo.61.1.6ayvl8t0aekf8vmy
Temple, S. 2019 'Word Clouds are lame', Medium. https://towardsdatascience.com/word-clouds-are-lame-263d9cbc49b7
Toepoel, V., Vermeeren, B. and Metin, B. 2019 'Smileys, stars, hearts, buttons, tiles or grids: influence of response format on substantive response, questionnaire experience and response time', Bulletin of Sociological Methodology/Bulletin de Méthodologie Sociologique 142(1), 57–74. https://doi.org/10.1177/0759106319834665
Tornes, A. 2021 'Enabling the future of academic research with the Twitter API', Twitter Developer Platform Blog. https://blog.twitter.com/developer/en_us/topics/tools/2021/enabling-the-future-of-academic-research-with-the-twitter-api
Twitter IR 2021 Q4 and Fiscal Year 2020: Letter to the Shareholders. https://s22.q4cdn.com/826641620/files/doc_financials/2020/q4/FINAL-Q4'20-TWTR-Shareholder-Letter.pdf
UK Government 2021 'COVID-19 response - Spring 2021 (summary)', GOV.UK, 22 February 2021. [Last accessed: 11 February 2022]. https://www.gov.uk/government/publications/covid-19-response-spring-2021/covid-19-response-spring-2021-summary
UNESCO 1972 Convention Concerning the Protection of the World Cultural and Natural Heritage, Paris. https://whc.unesco.org/archive/convention-en.pdf
UNESCO 1992 Convention Concerning the Protection of the World Cultural and Natural Heritage. World Heritage Committee, Santa-Fe, US. https://whc.unesco.org/archive/1992/whc-92-conf002-12e.pdf
UNESCO 1997 Convention Concerning the Protection of the World Cultural and Natural Heritage. World Heritage Committee, Naples, Italy. http://whc.unesco.org/archive/1997/whc-97-conf208-inf4e.pdf
UNESCO 2021 The Operational Guidelines for the Implementation of the World Heritage Convention. Paris. https://whc.unesco.org/en/guidelines/
UNESCO n.d. UNESCO World Heritage Centre - World Heritage List. https://whc.unesco.org/en/list/?search=cultural+landscape&themes=4&order=country
Unicode Consortium 2020 The Unicode Standard, Version 13.0, https://unicode.org/versions/Unicode13.0.0/
University of Edinburgh n.d. 'Digimap', Edina. https://digimap.edina.ac.uk/
University of Stirling n.d. 'Social value toolkit guidance for heritage practitioners', Social Value Toolkit. https://socialvalue.stir.ac.uk/
Vincent, J. 2021 'Facebook bans academics who researched ad transparency and misinformation on Facebook', The Verge. https://www.theverge.com/2021/8/4/22609020/facebook-bans-academic-researchers-ad-transparency-misinformation-nyu-ad-observatory-plug-in
Internet Archaeology is an open access journal based in the Department of Archaeology, University of York. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.
Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.