Johan Bollen

Twitter mood predicts the stock market

Behavioral economics tells us that emotions can profoundly affect individual behavior and decision-making. Does this also apply to societies at large, i.e. can societies experience mood states that affect their collective decision making? By extension is the public mood correlated or even predictive of economic indicators? Here we investigate whether measurements of collective mood states derived from large-scale Twitter feeds are correlated to the value of the Dow Jones Industrial Average (DJIA) over time. We analyze the text content of daily Twitter feeds by two mood tracking tools, namely OpinionFinder that measures positive vs. negative mood and Google-Profile of Mood States (GPOMS) that measures mood in terms of 6 dimensions (Calm, Alert, Sure, Vital, Kind, and Happy). We cross-validate the resulting mood time series by comparing their ability to detect the publics response to the presidential election and Thanksgiving day in 2008. A Granger causality analysis and a Self-Organizing Fuzzy Neural Network are then used to investigate the hypothesis that public mood states, as measured by the OpinionFinder and GPOMS mood time series, are predictive of changes in DJIA closing values. Our results indicate that the accuracy of DJIA predictions can be significantly improved by the inclusion of specific public mood dimensions but not others. We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the DJIA and a reduction of the Mean Average Percentage Error by more than 6%. Index Terms—stock market prediction — twitter — mood analysis.

Co-authorship networks in the digital library research community

The field of digital libraries (DLs) coalesced in 1994: the first digital library conferences were held that year, awareness of the World Wide Web was accelerating, and the National Science Foundation awarded

A Principal Component Analysis of 39 Scientific Impact Measures

24 Million (US) for the Digital Library Initiative (DLI). In this paper we examine the state of the DL domain after a decade of activity by applying social network analysis to the co-authorship network of the past ACM, IEEE, and joint ACM/IEEE digital library conferences. We base our analysis on a common binary undirectional network model to represent the co-authorship network, and from it we extract several established network measures. We also introduce a weighted directional network model to represent the co-authorship network, for which we define AuthorRank as an indicator of the impact of an individual author in the network. The results are validated against conference program committee members in the same period. The results show clear advantages of PageRank and AuthorRank over degree, closeness and betweenness centrality metrics. We also investigate the amount and nature of international participation in Joint Conference on Digital Libraries (JCDL).

Toward alternative metrics of journal impact: a comparison of download and citation data

Background The impact of scientific publications has traditionally been expressed in terms of citation counts. However, scientific activity has moved online over the past decade. To better capture scientific impact in the digital era, a variety of new impact measures has been proposed on the basis of social network analysis and usage log data. Here we investigate how these new measures relate to each other, and how accurately and completely they express scientific impact. Methodology We performed a principal component analysis of the rankings produced by 39 existing and proposed measures of scholarly impact that were calculated on the basis of both citation and usage log data. Conclusions Our results indicate that the notion of scientific impact is a multi-dimensional construct that can not be adequately measured by any single indicator, although some measures are more suitable than others. The commonly used citation Impact Factor is not positioned at the core of this construct, but at its periphery, and should thus be used with caution.

How the Scientific Community Reacts to Newly Submitted Preprints: Article Downloads, Twitter Mentions, and Citations

We generated networks of journal relationships from citation and download data, and determined journal impact rankings from these networks using a set of social network centrality metrics. The resulting journal impact rankings were compared to the ISI IF. Results indicate that, although social network metrics and ISI IF rankings deviate moderately for citation-based journal networks, they differ considerably for journal networks derived from download data. We believe the results represent a unique aspect of general journal impact that is not captured by the ISI IF. These results furthermore raise questions regarding the validity of the ISI IF as the sole assessment of journal impact, and suggest the possibility of devising impact metrics based on usage information in general.

Happiness is assortative in online social networks

We analyze the online response to the preprint publication of a cohort of 4,606 scientific articles submitted to the preprint database arXiv.org between October 2010 and May 2011. We study three forms of responses to these preprints: downloads on the arXiv.org site, mentions on the social media site Twitter, and early citations in the scholarly record. We perform two analyses. First, we analyze the delay and time span of article downloads and Twitter mentions following submission, to understand the temporal configuration of these reactions and whether one precedes or follows the other. Second, we run regression and correlation tests to investigate the relationship between Twitter mentions, arXiv downloads, and article citations. We find that Twitter mentions and arXiv downloads of scholarly articles follow two distinct temporal patterns of activity, with Twitter mentions having shorter delays and narrower time spans than arXiv downloads. We also find that the volume of Twitter mentions is statistically correlated with arXiv downloads and early citations just months after the publication of a preprint, with a possible bias that favors highly mentioned articles.

Clickstream data yields high-resolution maps of science

Online social networking communities may exhibit highly complex and adaptive collective behaviors. Since emotions play such an important role in human decision making, how online networks modulate human collective mood states has become a matter of considerable interest. In spite of the increasing societal importance of online social networks, it is unknown whether assortative mixing of psychological states takes place in situations where social ties are mediated solely by online networking services in the absence of physical contact. Here, we show that the general happiness, or subjective well-being (SWB), of Twitter users, as measured from a 6-month record of their individual tweets, is indeed assortative across the Twitter social network. Our results imply that online social networks may be equally subject to the social mechanisms that cause assortative mixing in real social networks and that such assortative mixing takes place at the level of SWB. Given the increasing prevalence of online social networks, their propensity to connect users with similar levels of SWB may be an important factor in how positive and negative sentiments are maintained and spread through human society. Future research may focus on how event-specific mood states can propagate and influence user behavior in “real life.”

More Tweets, More Votes: Social Media as a Quantitative Indicator of Political Behavior

Background Intricate maps of science have been created from citation data to visualize the structure of scientific activity. However, most scientific publications are now accessed online. Scholarly web portals record detailed log data at a scale that exceeds the number of all existing citations combined. Such log data is recorded immediately upon publication and keeps track of the sequences of user requests (clickstreams) that are issued by a variety of users across many different domains. Given these advantages of log datasets over citation data, we investigate whether they can produce high-resolution, more current maps of science. Methodology Over the course of 2007 and 2008, we collected nearly 1 billion user interactions recorded by the scholarly web portals of some of the most significant publishers, aggregators and institutional consortia. The resulting reference data set covers a significant part of world-wide use of scholarly web portals in 2006, and provides a balanced coverage of the humanities, social sciences, and natural sciences. A journal clickstream model, i.e. a first-order Markov chain, was extracted from the sequences of user interactions in the logs. The clickstream model was validated by comparing it to the Getty Research Institutes Architecture and Art Thesaurus. The resulting model was visualized as a journal network that outlines the relationships between various scientific domains and clarifies the connection of the social sciences and humanities to the natural sciences. Conclusions Maps of science resulting from large-scale clickstream data provide a detailed, contemporary view of scientific activity and correct the underrepresentation of the social sciences and humanities that is commonly found in citation data.

Twitter Mood as a Stock Market Predictor

Is social media a valid indicator of political behavior? There is considerable debate about the validity of data extracted from social media for studying offline behavior. To address this issue, we show that there is a statistically significant association between tweets that mention a candidate for the U.S. House of Representatives and his or her subsequent electoral performance. We demonstrate this result with an analysis of 542,969 tweets mentioning candidates selected from a random sample of 3,570,054,618, as well as Federal Election Commission data from 795 competitive races in the 2010 and 2012 U.S. congressional elections. This finding persists even when controlling for incumbency, district partisanship, media coverage of the race, time, and demographic variables such as the districts racial and gender composition. Our findings show that reliable data about political behavior can be extracted from social media.

Usage bibliometrics

Behavioral finance researchers can apply computational methods to large-scale social media data to better understand and predict markets.