Using Semantic Similarity and Text Embedding to Measure the Social Media Echo of Strategic Communications

Online discourse covers a wide range of topics and many actors tailor their content to impact online discussions through carefully crafted messages and targeted campaigns. Yet the scale and diversity of online media content make it difficult to evaluate the impact of a particular message. In this paper, we present a new technique that leverages semantic similarity to quantify the change in the discussion after a particular message has been published. We use a set of press releases from environmental organisations and tweets from the climate change debate to show that our novel approach reveals a heavy-tailed distribution of response in online discourse to strategic communications.

[1]  Hywel T. P. Williams,et al.  Sponsored messaging about climate change on Facebook: Actors, content, frames , 2022, 2211.13965.

[2]  J. Bryden,et al.  Does United Kingdom parliamentary attention follow social media posts? , 2022, EPJ Data Science.

[3]  Przemyslaw K. Joniak,et al.  Gender Biases and Where to Find Them: Exploring Gender Bias in Pre-Trained Transformer-based Language Models Using Movement Pruning , 2022, GEBNLP.

[4]  B. Nyhan,et al.  Time and skeptical opinion content erode the effects of science coverage on climate beliefs and attitudes , 2022, Proceedings of the National Academy of Sciences of the United States of America.

[5]  Yulia Tsvetkov,et al.  A Survey of Race, Racism, and Anti-Racism in NLP , 2021, ACL.

[6]  Matthew H. Goldberg,et al.  Shifting Republican views on climate change through targeted advertising , 2021, Nature Climate Change.

[7]  Dave Dopson,et al.  Fast WordPiece Tokenization , 2020, EMNLP.

[8]  Natalia Criado,et al.  Discovering and Categorising Language Biases in Reddit , 2020, ICWSM.

[9]  Kathleen M. Carley,et al.  Characterizing communities of hashtag usage on twitter during the 2020 COVID-19 pandemic by multi-view clustering , 2020, Appl. Netw. Sci..

[10]  Jacob N. Shapiro,et al.  Content-based features predict social media influence operations , 2020, Science Advances.

[11]  Jianfeng Gao,et al.  UniLMv2: Pseudo-Masked Language Models for Unified Language Model Pre-Training , 2020, ICML.

[12]  Iryna Gurevych,et al.  Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks , 2019, EMNLP.

[13]  Jay Jennings,et al.  Mobilizing Millennial Voters with Targeted Internet Advertisements: A Field Experiment , 2019, Political Communication.

[14]  Stefanie Haustein,et al.  Identifying diffusion patterns of research articles on Twitter: A case study of online engagement with open access articles , 2019, Public understanding of science.

[15]  W. Pearce,et al.  The social media life of climate change: Platforms, publics, and future imaginaries , 2018, WIREs Climate Change.

[16]  Jessica T. Feezell.  Agenda Setting through Social Media: The Importance of Incidental News Exposure and Social Filtering in the Digital Era , 2018 .

[17]  Luis F. Chiroque,et al.  Unsupervised Scalable Statistical Method for Identifying Influential Users in Online Social Networks , 2018, Scientific Reports.

[18]  P. Howe,et al.  Experimental effects of climate messages vary geographically , 2018, Nature Climate Change.

[19]  Arnaud Legout,et al.  Social Clicks: What and Who Gets Read on Twitter? , 2016, SIGMETRICS.

[20]  Shlomo Havlin,et al.  Local structure can identify and quantify influential global spreaders in large scale social networks , 2015, Proceedings of the National Academy of Sciences.

[21]  Filippo Menczer,et al.  Clustering memes in social media streams , 2014, Social Network Analysis and Mining.

[22]  Filippo Menczer,et al.  Predicting Successful Memes Using Network and Community Structure , 2014, ICWSM.

[23]  Jürgen Pfeffer,et al.  Characterizing the life cycle of online news stories using social media reactions , 2013, CSCW.

[24]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[25]  Wiebke Wagner,et al.  Steven Bird, Ewan Klein and Edward Loper: Natural Language Processing with Python, Analyzing Text with the Natural Language Toolkit , 2010, Lang. Resour. Evaluation.

[26]  A. Nederhof.  Methods of coping with social desirability bias: A review. , 1985 .