An iterative topic model filtering framework for short and noisy user-generated data: analyzing conspiracy theories on twitter

Kant, Gillian; Wiebelt, Levin; Weisser, Christoph; Kis‐Katos, Krisztina; Luber, Mattias; Säfken, Benjamin

doi:10.1007/s41060-022-00321-4

Cited by 8 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When they use unconventional hashtags, they go 'under the radar.' A different datacollection strategy should have been employed, were we looking for such tweets, as described for example in [19][20][21][22].…”

Section: Restrictionsmentioning

confidence: 99%

Emotions during the Pandemic’s First Wave: The Case of Greek Tweets

Skarpelos,

Messini,

Roinioti

et al. 2024

Digital

View full text Add to dashboard Cite

While most published research on COVID-19 focused on a few countries and especially on the second wave of the pandemic and the vaccination period, we turn to the first wave (March–May 2020) to examine the sentiments and emotions expressed by Twitter users in Greece. Using deep-learning techniques, the analysis reveals a complex interplay of surprise, anger, fear, and sadness. Initially, surprise was dominant, reflecting the shock and uncertainty accompanying the sudden onset of the pandemic. Anger replaced surprise as individuals struggled with isolation and social distancing. Despite these challenges, positive sentiments of hope, resilience and solidarity were also expressed. The COVID-19 pandemic had a strong imprint upon the emotional landscape worldwide and in Greece. This calls for appealing to emotions as well as to reason when crafting effective public health strategies.

show abstract

Section: Restrictionsmentioning

confidence: 99%

Emotions during the Pandemic’s First Wave: The Case of Greek Tweets

Skarpelos,

Messini,

Roinioti

et al. 2024

Digital

View full text Add to dashboard Cite

show abstract

“…When they use unconventional hashtags, they go 'under the radar.' A different data-collection strategy should be employed, were we looking for such tweets, as described for example in [19][20][21][22].…”

Section: Restrictionsmentioning

confidence: 99%

Emotions during the Pandemic’s First Wave: The Case of Greek Tweets

Skarpelos,

Messini,

Roinioti

et al. 2023

Preprint

View full text Add to dashboard Cite

While most published research on COVID-19 focused in a few countries and especially during the second wave of the pandemic and the vaccination period, we turn to the first wave (march-May 2020) to examine sentiment and emotions expressed by Twitter users in Greece. We use computational methods combining opinion mining and the application of a modified emotion lexicon with Social Network Analysis to explore the subtleties behind an overarching negative sentiment. Our analysis has shown that popular sentiment isn’t stable nor monolithic, since underlying emotions may be a better predictor of attitudes during a crisis.

show abstract

“…However, these techniques are also very popular in the social sciences, with applications such as text analysis of conspiracy theories on Twitter (Kant et al, 2022) or the prediction of economic variables such as stock prices (Thormann et al, 2021). But only recently has attention focused on an integral part of scientific applications: model selection, interpretation and especially model evaluation.One of the focal points of the current research is the attempt to achieve a better explainability of the deep learning models and the associated evaluation of the models.…”

Section: Introductionmentioning

confidence: 99%

Estimating and evaluating mixed and semiparametric models with statistical and deep learning methods

Kruse

View full text Add to dashboard Cite

Semiparametric models are well-established, versatile and effective statistical models for analysing complex data by combining both parametric and non-parametric components and are used in a wide range of applications and fields. Meanwhile, deep learning models are a type of machine learning model designed to identify and represent intricate patterns and relationships in data, which are gaining popularity due to their impressive performance in various applications, including analysing structured tabular data and, in contrast to statistical models, unstructured data such as images, text, and sound recordings. However, the complexity of these models can make them challenging to understand and interpret, limiting their transparency and interpretability.The purpose of this thesis is twofold. Firstly, develop more efficient model selection and evaluation methods that can handle model uncertainty for semiparametric and deep learning models.Secondly, to merge traditional statistical methods with machine and deep learning concepts, combining their strengths to mitigate their weaknesses.Based on Stein's unbiased risk estimate, part one of this thesis introduces a criterion for determining squared loss optimal weights for model averaging of (conditional) linear mixed models. Furthermore, the complicated underlying optimisation of the presented criterion is discussed, and a possible solution via a specifically customised algorithm based on the augmented Lagrangian is introduced.An essential part of model evaluation and selection in statistics is model complexity, often measured in degrees of freedom. In the second part of this thesis, three different methods for measuring model complexity based on the concept of covariance penalties associated with degrees of freedom are presented and ultimately compared and analysed using different simulations.The third and final section of the thesis introduces a new type of neural additive models for location, scale, and shape by combining the GAMLSS distribution regression framework with neural network techniques and principles. This approach differs from previous deep learning methods as it can model the entire response distribution rather than just the mean response. The effectiveness of this method is evaluated using simulated and real data and compared against established statistical and deep learning methodologies. iii Zusammenfassung Semiparametrische Modelle sind bewährte, vielseitige und effiziente statistische Modelle für die Analyse komplexer Daten, die parametrische und nichtparametrische Komponenten kombinieren und in einer Vielzahl von Anwendungen und Bereichen eingesetzt werden. Deep Learning Modelle sind eine Art Machine Learning-Modelle, die darauf ausgelegt sind, komplexe Muster und Beziehungen in Daten zu erkennen und darzustellen. Sie erfreuen sich aufgrund ihrer beeindruckenden Leistungsfähigkeit in verschiedenen Anwendungen immer größerer Beliebtheit, z.B. bei der Analyse von strukturierten Tabellendaten oder, im Gegensatz zu statistischen Modellen, von unstrukturie...

show abstract

An iterative topic model filtering framework for short and noisy user-generated data: analyzing conspiracy theories on twitter

Cited by 8 publications

References 25 publications

Emotions during the Pandemic’s First Wave: The Case of Greek Tweets

Emotions during the Pandemic’s First Wave: The Case of Greek Tweets

Emotions during the Pandemic’s First Wave: The Case of Greek Tweets

Estimating and evaluating mixed and semiparametric models with statistical and deep learning methods

Contact Info

Product

Resources

About