arXiv subjects

Eni Mustafaraj

Publications and source records attributed to Eni Mustafaraj.

5 recordsLinked to original sources

arXiv2025

Creating Targeted, Interpretable Topic Models with LLM-Generated Text Augmentation

Unsupervised machine learning techniques, such as topic modeling and clustering, are often used to identify latent patterns in unstructured text data in fields such as political science and sociology. These methods overcome common concerns about reproducibility and costliness involved in the labor-intensive process of human qualitative analysis. However, two major limitations of topic models are their interpretability and their practicality for answering targeted, domain-specific social science research questions. In this work, we investigate opportunities for using LLM-generated text augmentation to improve the usefulness of topic modeling output. We use a political science case study to evaluate our results in a domain-specific application, and find that topic modeling using GPT-4 augmentations creates highly interpretable categories that can be used to investigate domain-specific research questions with minimal human guidance.

cs.CL↗

arXiv2024

Algorithmic Misjudgement in Google Search Results: Evidence from Auditing the US Online Electoral Information Environment

Google Search is an important way that people seek information about politics, and Google states that it is ``committed to providing timely and authoritative information on Google Search to help voters understand, navigate, and participate in democratic processes.'' This paper studies the extent to which government-maintained web domains are represented in the online electoral information environment, as captured through 3.45 Google Search result pages collected during the 2022 US midterm elections for 786 locations across the United States. Focusing on state, county, and local government domains that provide locality-specific information, we study not only the extent to which these sources appear in organic search results, but also the extent to which these sources are correctly targeted to their respective constituents. We label misalignment between the geographic area that non-federal domains serve and the locations for which they appear in search results as algorithmic mistargeting, a subtype of algorithmic misjudgement in which the search algorithm targets locality-specific information to users in different (incorrect) locations. In the context of the 2022 US midterm elections, we find that 71% of all occurrences of state, county, and local government sources were mistargeted, with some domains appearing disproportionately often among organic results despite providing locality-specific information that may not be relevant to all voters. However, we also find that mistargeting often occurs in low ranks. We conclude by considering the potential consequences of extensive mistargeting of non-federal government sources and argue that ensuring the correct targeting of these sources to their respective constituents is a critical part of Google's role in facilitating access to authoritative and locally-relevant electoral information.

cs.CY↗

arXiv2017

The Fake News Spreading Plague: Was it Preventable?

In 2010, a paper entitled "From Obscurity to Prominence in Minutes: Political Speech and Real-time search" won the Best Paper Prize of the Web Science 2010 Conference. Among its findings were the discovery and documentation of what was termed a "Twitter-bomb", an organized effort to spread misinformation about the democratic candidate Martha Coakley through anonymous Twitter accounts. In this paper, after summarizing the details of that event, we outline the recipe of how social networks are used to spread misinformation. One of the most important steps in such a recipe is the "infiltration" of a community of users who are already engaged in conversations about a topic, to use them as organic spreaders of misinformation in their extended subnetworks. Then, we take this misinformation spreading recipe and indicate how it was successfully used to spread fake news during the 2016 U.S. Presidential Election. The main differences between the scenarios are the use of Facebook instead of Twitter, and the respective motivations (in 2010: political influence; in 2016: financial benefit through online advertising). After situating these events in the broader context of exploiting the Web, we seize this opportunity to address limitations of the reach of research findings and to start a conversation about how communities of researchers can increase their impact on real-world societal issues.

cs.SI↗

arXiv2014

Investigating Rumor Propagation with TwitterTrails

Social media have become part of modern news reporting, used by journalists to spread information and find sources, or as a news source by individuals. The quest for prominence and recognition on social media sites like Twitter can sometimes eclipse accuracy and lead to the spread of false information. As a way to study and react to this trend, we introduce {\sc TwitterTrails}, an interactive, web-based tool ({\tt twittertrails.com}) that allows users to investigate the origin and propagation characteristics of a rumor and its refutation, if any, on Twitter. Visualizations of burst activity, propagation timeline, retweet and co-retweeted networks help its users trace the spread of a story. Within minutes {\sc TwitterTrails} will collect relevant tweets and automatically answer several important questions regarding a rumor: its originator, burst characteristics, propagators and main actors according to the audience. In addition, it will compute and report the rumor's level of visibility and, as an example of the power of crowdsourcing, the audience's skepticism towards it which correlates with the rumor's credibility. We envision {\sc TwitterTrails} as valuable tool for individual use, but we especially for amateur and professional journalists investigating recent and breaking stories. Further, its expanding collection of investigated rumors can be used to answer questions regarding the amount and success of misinformation on Twitter.

cs.SI↗

arXiv2014

Do Retweets indicate Interest, Trust, Agreement? (Extended Abstract)

Arguably one of the most important features of Twitter is the support for "retweets" or messages re-posted verbatim by a user that were originated by someone else. (This does not include modified tweets that sometimes are referred to as retweets.) Despite the fact that retweets are routinely studied and reported, many important questions remain about user motivation for their use and their significance. In this paper we answer the question of what users indicate when they retweet. We do so in a comprehensive fashion, by employing a user survey, a study of user profiles, and a meta-analysis of over 100 research publications from three related major conferences. Our findings indicate that retweeting indicates not only interest in a message, but also trust in the message and the originator, and agreement with the message contents. However, the findings are significantly weaker for journalists, some of whom beg to differ declaring so in their own user profiles. On the other hand, the inclusion of hashtags strengthens the signal of agreement, especially when the hashtags are related to politics. While in the past there have been additional claims in the literature about possible reasons for retweeting, many of them are not supported, especially given the technical changes introduced recently by Twitter.

cs.SI↗