Search

Recent studies suggest social media activity can function as a proxy for measures of state-level public health, detectable through natural language processing. We present results of our efforts to apply this approach to estimate homelessness at the state level throughout the US during the period 2010-2019 and 2022 using a dataset of roughly 1 million geotagged tweets containing the substring ``homeless.'' Correlations between homelessness-related tweet counts and ranked per capita homelessness volume, but not general-population densities, suggest a relationship between the likelihood of Twitter users to personally encounter or observe homelessness in their everyday lives and their likelihood to communicate about it online. An increase to the log-odds of ``homeless'' appearing in an English-language tweet, as well as an acceleration in the increase in average tweet sentiment, suggest that tweets about homelessness are also affected by trends at the nation-scale. Additionally, changes to the lexical content of tweets over time suggest that reversals to the polarity of national or state-level trends may be detectable through an increase in political or service-sector language over the semantics of charity or direct appeals. An analysis of user account type also revealed changes to Twitter-use patterns by accounts authored by individuals versus entities that may provide an additional signal to confirm changes to homelessness density in a given jurisdiction. While a computational approach to social media analysis may provide a low-cost, real-time dataset rich with information about nationwide and localized impacts of homelessness and homelessness policy, we find that practical issues abound, limiting the potential of social media as a proxy to complement other measures of homelessness., Comment: 29 pages, 21 figures

Well curated, large-scale corpora of social media posts containing broad public opinion offer an alternative data source to complement traditional surveys. While surveys are effective at collecting representative samples and are capable of achieving high accuracy, they can be both expensive to run and lag public opinion by days or weeks. Both of these drawbacks could be overcome with a real-time, high volume data stream and fast analysis pipeline. A central challenge in orchestrating such a data pipeline is devising an effective method for rapidly selecting the best corpus of relevant documents for analysis. Querying with keywords alone often includes irrelevant documents that are not easily disambiguated with bag-of-words natural language processing methods. Here, we explore methods of corpus curation to filter irrelevant tweets using pre-trained transformer-based models, fine-tuned for our binary classification task on hand-labeled tweets. We are able to achieve F1 scores of up to 0.95. The low cost and high performance of fine-tuning such a model suggests that our approach could be of broad benefit as a pre-processing step for social media datasets with uncertain corpus boundaries., Comment: 12 pages, 6 figures

We explore the relationship between context and happiness scores in political tweets using word co-occurrence networks, where nodes in the network are the words, and the weight of an edge is the number of tweets in the corpus for which the two connected words co-occur. In particular, we consider tweets with hashtags #imwithher and #crookedhillary, both relating to Hillary Clinton's presidential bid in 2016. We then analyze the network properties in conjunction with the word scores by comparing with null models to separate the effects of the network structure and the score distribution. Neutral words are found to be dominant and most words, regardless of polarity, tend to co-occur with neutral words. We do not observe any score homophily among positive and negative words. However, when we perform network backboning, community detection results in word groupings with meaningful narratives, and the happiness scores of the words in each group correspond to its respective theme. Thus, although we observe no clear relationship between happiness scores and co-occurrence at the node or edge level, a community-centric approach can isolate themes of competing sentiments in a corpus.

Sentiment-aware intelligent systems are essential to a wide array of applications. These systems are driven by language models which broadly fall into two paradigms: Lexicon-based and contextual. Although recent contextual models are increasingly dominant, we still see demand for lexicon-based models because of their interpretability and ease of use. For example, lexicon-based models allow researchers to readily determine which words and phrases contribute most to a change in measured sentiment. A challenge for any lexicon-based approach is that the lexicon needs to be routinely expanded with new words and expressions. Here, we propose two models for automatic lexicon expansion. Our first model establishes a baseline employing a simple and shallow neural network initialized with pre-trained word embeddings using a non-contextual approach. Our second model improves upon our baseline, featuring a deep Transformer-based network that brings to bear word definitions to estimate their lexical polarity. Our evaluation shows that both models are able to score new words with a similar accuracy to reviewers from Amazon Mechanical Turk, but at a fraction of the cost., Comment: 17 pages, 8 figures

The murder of George Floyd by police in May 2020 sparked international protests and renewed attention in the Black Lives Matter movement. Here, we characterize ways in which the online activity following George Floyd's death was unparalleled in its volume and intensity, including setting records for activity on Twitter, prompting the saddest day in the platform's history, and causing George Floyd's name to appear among the ten most frequently used phrases in a day, where he is the only individual to have ever received that level of attention who was not known to the public earlier that same week. Further, we find this attention extended beyond George Floyd and that more Black victims of fatal police violence received attention following his death than during other past moments in Black Lives Matter's history. We place that attention within the context of prior online racial justice activism by showing how the names of Black victims of police violence have been lifted and memorialized over the last 12 years on Twitter. Our results suggest that the 2020 wave of attention to the Black Lives Matter movement centered past instances of police violence in an unprecedented way, demonstrating the impact of the movement's rhetorical strategy to "say their names."

Mental health challenges are thought to afflict around 10% of the global population each year, with many going untreated due to stigma and limited access to services. Here, we explore trends in words and phrases related to mental health through a collection of 1- , 2-, and 3-grams parsed from a data stream of roughly 10% of all English tweets since 2012. We examine temporal dynamics of mental health language, finding that the popularity of the phrase 'mental health' increased by nearly two orders of magnitude between 2012 and 2018. We observe that mentions of 'mental health' spike annually and reliably due to mental health awareness campaigns, as well as unpredictably in response to mass shootings, celebrities dying by suicide, and popular fictional stories portraying suicide. We find that the level of positivity of messages containing 'mental health', while stable through the growth period, has declined recently. Finally, we use the ratio of original tweets to retweets to quantify the fraction of appearances of mental health language due to social amplification. Since 2015, mentions of mental health have become increasingly due to retweets, suggesting that stigma associated with discussion of mental health on Twitter has diminished with time., Comment: 12 pages, 5 figures, 1 table

In real-time, social media data strongly imprints world events, popular culture, and day-to-day conversations by millions of ordinary people at a scale that is scarcely conventionalized and recorded. Vitally, and absent from many standard corpora such as books and news archives, sharing and commenting mechanisms are native to social media platforms, enabling us to quantify social amplification (i.e., popularity) of trending storylines and contemporary cultural phenomena. Here, we describe Storywrangler, a natural language processing instrument designed to carry out an ongoing, day-scale curation of over 100 billion tweets containing roughly 1 trillion 1-grams from 2008 to 2021. For each day, we break tweets into unigrams, bigrams, and trigrams spanning over 100 languages. We track n-gram usage frequencies, and generate Zipf distributions, for words, hashtags, handles, numerals, symbols, and emojis. We make the data set available through an interactive time series viewer, and as downloadable time series and daily distributions. Although Storywrangler leverages Twitter data, our method of extracting and tracking dynamic changes of n-grams can be extended to any similar social media platform. We showcase a few examples of the many possible avenues of study we aim to enable including how social amplification can be visualized through 'contagiograms'. We also present some example case studies that bridge n-gram time series with disparate data sources to explore sociotechnical dynamics of famous individuals, box office success, and social unrest., Comment: Main text: 15 pages, 6 figures; Supplementary text: 23 pages, 11 figures, 15 tables. Website: https://storywrangling.org/

The past decade has witnessed a marked increase in the use of social media by politicians, most notably exemplified by the 45th President of the United States (POTUS), Donald Trump. On Twitter, POTUS messages consistently attract high levels of engagement as measured by likes, retweets, and replies. Here, we quantify the balance of these activities, also known as "ratios", and study their dynamics as a proxy for collective political engagement in response to presidential communications. We find that raw activity counts increase during the period leading up to the 2016 election, accompanied by a regime change in the ratio of retweets-to-replies connected to the transition between campaigning and governing. For the Trump account, we find words related to fake news and the Mueller inquiry are more common in tweets with a high number of replies relative to retweets. Finally, we find that Barack Obama consistently received a higher retweet-to-reply ratio than Donald Trump. These results suggest Trump's Twitter posts are more often controversial and subject to enduring engagement as a given news cycle unfolds., Comment: 17 pages, 10 figures

Using a random 10% sample of tweets authored from 2019-09-01 through 2020-04-30, we analyze the dynamic behavior of words (1-grams) used on Twitter to describe the ongoing COVID-19 pandemic. Across 24 languages, we find two distinct dynamic regimes: One characterizing the rise and subsequent collapse in collective attention to the initial Coronavirus outbreak in late January, and a second that represents March COVID-19-related discourse. Aggregating countries by dominant language use, we find that volatility in the first dynamic regime is associated with future volatility in new cases of COVID-19 roughly three weeks (average 22.49 $\pm$ 3.26 days) later. Our results suggest that surveillance of change in usage of epidemiology-related words on social media may be useful in forecasting later change in disease case numbers, but we emphasize that our current findings are not causal or necessarily predictive., Comment: 12 + 4 pages, 11 + 4 figures, code + data + figures will soon be available at http://compstorylab.org/covid19ngrams/

We study collective attention paid towards hurricanes through the lens of $n$-grams on Twitter, a social media platform with global reach. Using hurricane name mentions as a proxy for awareness, we find that the exogenous temporal dynamics are remarkably similar across storms, but that overall collective attention varies widely even among storms causing comparable deaths and damage. We construct `hurricane attention maps' and observe that hurricanes causing deaths on (or economic damage to) the continental United States generate substantially more attention in English language tweets than those that do not. We find that a hurricane's Saffir-Simpson wind scale category assignment is strongly associated with the amount of attention it receives. Higher category storms receive higher proportional increases of attention per proportional increases in number of deaths or dollars of damage, than lower category storms. The most damaging and deadly storms of the 2010s, Hurricanes Harvey and Maria, generated the most attention and were remembered the longest, respectively. On average, a category 5 storm receives 4.6 times more attention than a category 1 storm causing the same number of deaths and economic damage., Comment: 31 pages (14 main, 17 Supplemental), 19 figures (5 main, 14 appendix)

Working from a dataset of 118 billion messages running from the start of 2009 to the end of 2019, we identify and explore the relative daily use of over 150 languages on Twitter. We find that eight languages comprise 80% of all tweets, with English, Japanese, Spanish, and Portuguese being the most dominant. To quantify social spreading in each language over time, we compute the 'contagion ratio': The balance of retweets to organic messages. We find that for the most common languages on Twitter there is a growing tendency, though not universal, to retweet rather than share new content. By the end of 2019, the contagion ratios for half of the top 30 languages, including English and Spanish, had reached above 1 -- the naive contagion threshold. In 2019, the top 5 languages with the highest average daily ratios were, in order, Thai (7.3), Hindi, Tamil, Urdu, and Catalan, while the bottom 5 were Russian, Swedish, Esperanto, Cebuano, and Finnish (0.26). Further, we show that over time, the contagion ratios for most common languages are growing more strongly than those of rare languages., Comment: 26 pages (15 main, 11 appendix), 13 figures (6 main, 7 appendix), and 4 online appendices available at http://compstorylab.org/storywrangler/papers/tlid/

When building a global brand of any kind -- a political actor, clothing style, or belief system -- developing widespread awareness is a primary goal. Short of knowing any of the stories or products of a brand, being talked about in whatever fashion -- raw fame -- is, as Oscar Wilde would have it, better than not being talked about at all. Here, we measure, examine, and contrast the day-to-day raw fame dynamics on Twitter for US Presidents and major US Presidential candidates from 2008 to 2020: Barack Obama, John McCain, Mitt Romney, Hillary Clinton, Donald Trump, and Joe Biden. We assign ``lexical fame'' to be the number and (Zipfian) rank of the (lowercased) mentions made for each individual across all languages. We show that all five political figures have at some point reached extraordinary volume levels of what we define to be ``lexical ultrafame'': An overall rank of approximately 300 or less which is largely the realm of function words and demarcated by the highly stable rank of `god'. By this measure, `trump' has become enduringly ultrafamous, from the 2016 election on. We use typical ranks for country names and function words as standards to improve perception of scale. We quantify relative fame rates and find that in the eight weeks leading up the 2008 and 2012 elections, `obama' held a 1000:757 volume ratio over `mccain' and 1000:892 over `romney', well short of the 1000:544 and 1000:504 volumes favoring `trump' over `hillary' and `biden' in the 8 weeks leading up to the 2016 and 2020 elections. Finally, we track how only one other entity has more sustained ultrafame than `trump' on Twitter: The K-pop (Korean pop) band BTS. We chart the dramatic rise of BTS, finding their Twitter handle `@bts\_twt' has been able to compete with `a' and `the'. Our findings for BTS more generally point to K-pop's growing economic, social, and political power., Comment: 31 pages (21 pages main text, 10 pages appendix), 8 figures (7 in main text, 1 in appendix), 10 tables (1 in main text, 9 in appendix)

We introduce a qualitative, shape-based, timescale-independent time-domain transform used to extract local dynamics from sociotechnical time series---termed the Discrete Shocklet Transform (DST)---and an associated similarity search routine, the Shocklet Transform And Ranking (STAR) algorithm, that indicates time windows during which panels of time series display qualitatively-similar anomalous behavior. After distinguishing our algorithms from other methods used in anomaly detection and time series similarity search, such as the matrix profile, seasonal-hybrid ESD, and discrete wavelet transform-based procedures, we demonstrate the DST's ability to identify mechanism-driven dynamics at a wide range of timescales and its relative insensitivity to functional parameterization. As an application, we analyze a sociotechnical data source (usage frequencies for a subset of words on Twitter) and highlight our algorithms' utility by using them to extract both a typology of mechanistic local dynamics and a data-driven narrative of socially-important events as perceived by English-language Twitter., Comment: 29 pages (20 body, 9 appendix), 20 figures (13 body, 7 appendix), three online appendices available at http://compstorylab.org/shocklets/ (two displaying interactive visualizations and one containing over 10,000 figures), open-source implementation of STAR algorithm and discrete shocklet transform available at https://gitlab.com/compstorylab/discrete-shocklet-transform