Dissecting Twitter Discussion Threads with Topic-Aware Network Visualization

Babvey, Pouria; Lipizzi, Carlo; Ramírez-Márquez, José Emmanuel

doi:10.1109/csci49370.2019.00254

Cited by 6 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Twitter is, in essence, a conversational social media communication tool that lends itself to be used by researchers as a data repository to understand humans' perspectives on different social issues. Given its pervasive use (at least in the US), Twitter can keep pace with fast‐moving and multiple series of events and at the same time facilitate opinion exchange by users 19 …”

Section: Methodsmentioning

confidence: 99%

“…Given its pervasive use (at least in the US), Twitter can keep pace with fast-moving and multiple series of events and at the same time facilitate opinion exchange by users. 19 During the COVID-19 pandemic 12 and other crises, 20,21 Twitter has been helpful to understand the behavior and thoughts of people. Several studies use Twitter to better understand humans' perspectives toward the new norms that are a result of the COVID-19 pandemic 22 perspectives on remote work during and after the initial months of COVID-19 remain unexplored.…”

Section: Twittermentioning

confidence: 99%

See 1 more Smart Citation

Impacts of work‐at‐home policies on systems engineers and the general population

Perez‐Pereda

Babvey

Ramírez-Márquez

et al. 2022

Systems Engineering

Self Cite

View full text Add to dashboard Cite

The COVID‐19 pandemic presented many challenges, one of them being the imposition of “work‐at‐home” policies in March 2020. The Systems Engineering Research Center (SERC) and the International Council on Systems Engineering (INCOSE) conducted two online surveys—one during the first months of the pandemic in 2020 and the second survey 1 year after, in March 2021—to understand the impact of these policies within the systems engineering community. The surveys' format consisted of multiple‐choice questions and open‐answer questions, which were analyzed using LDA for topic modeling. Data were also collected from social media during the same timeframes to compare the feelings and experiences of systems engineers with those of the general population.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Twittermentioning

confidence: 99%

Impacts of work‐at‐home policies on systems engineers and the general population

Perez‐Pereda

Babvey

Ramírez-Márquez

et al. 2022

Systems Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…Guo et al present a similar approach focusing on topic links that improve performance in removing irrelevant links (Guo et al 2015). Similar to these works, Babvey et al use topics to create "topic-aware" networks to analyze the nature of online conversations (Babvey, Lipizzi, and Ramirez-Marquez 2019). In this work, social media posts were the nodes, and they were colored by their assigned topic.…”

Section: Topic Modeling and Networkmentioning

confidence: 99%

“…There have been several recent works on conversation modeling in social media. The methods in these works often take both the conversation structure (e.g., reply tree in Twitter) and the textual content of the posts of the conversation to model social media conversations (Babvey, Lipizzi, and Ramirez-Marquez 2019;Mendoza, Parra, and Soto 2020;Benslimane et al 2023). The task of conversation modeling is often done for predicting contentious or controversial conversations (Babvey, Lipizzi, and Ramirez-Marquez 2019).…”

Section: Narrative and Conversation Networkmentioning

confidence: 99%

A Cross-Platform Topic Analysis of the Nazi Narrative on Twitter and Telegram during the 2022 Russian Invasion of Ukraine

Kloo,

Cruickshank,

Carley

2024

ICWSM

View full text Add to dashboard Cite

To influence the information landscape preceding and during the military invasion of Ukraine in February 2022, Russia initiated a disinformation campaign portraying Ukraine as a Nazi state. This study aims to compare discussions related to this campaign on Twitter and Telegram. The analysis reveals that the Nazis and Ukraine narrative was constant on Twitter but only emerged on Telegram after the invasion in channels that had previously focused on a broader set of conspiracy theories. Beyond the examination of Russian disinformation in this case study, the paper introduces an innovative methodology for constructing topic networks from social media data. This approach expands upon traditional topic modeling by incorporating the network properties of social media data to establish directed networks that characterize the interplay between conversation topics. Through this methodology, we gain the ability to observe topical evolutions, providing fresh insights into the disinformation campaign and its efficacy in shaping discussions around the Russian invasion on social media.

show abstract

“…BERT benefits from two main features: (1) Pretraining: different forms of information about the semantic and syntactic of the language is encoded in the model (Jawahar et al, 2019), and this let fine-tuning the model on a much smaller dataset than it would be required for a model that is built from the ground up. (2) multi-head attention layers: attention mechanism was first introduced in (Bahdanau et al, 2014) to focus on the most pertinent parts of the text for language translation and found applications in a wide range of NLP tasks (He et al, 2017;Babvey et al, 2019). Then, multi-head Figure 1: Consistency check for some sample outputs attention mechanism (Vaswani et al, 2017) initiated a new epoch in NLP by allowing attention heads to focus on different aspects of the language concurrently.…”

Section: Introductionmentioning

confidence: 99%

Pheonix at SemEval-2020 Task 5: Masking the Labels Lubricates Models for Sequence Labeling

Babvey

Borrelli

Zhao

et al. 2020

Proceedings of the Fourteenth Workshop on Semantic Evaluation

Self Cite

View full text Add to dashboard Cite

This paper presents the deep-learning model that is submitted to the SemEval-2020 Task 5 competition: "Detecting Counterfactuals". We participated in both Subtask1 and Subtask2. The model proposed in this paper ranked 2nd in Subtask2: "Detecting antecedent and consequence". Our model approaches the task as a sequence labeling. The architecture is built on top of BERT; and a multi-head attention layer with label masking is used to benefit from the mutual information between nearby labels. Also, for prediction, a multi-stage algorithm is used in which the model finalize some predictions with higher certainty in each step and use them in the following. Our results show that masking the labels not only is an efficient regularization method but also improves the accuracy of the model compared with other alternatives like CRF. Label masking can be used as a regularization method in sequence labeling. Also, it improves the performance of the model by learning the specific patterns in the target variable.

show abstract

Dissecting Twitter Discussion Threads with Topic-Aware Network Visualization

Cited by 6 publications

References 33 publications

Impacts of work‐at‐home policies on systems engineers and the general population

Impacts of work‐at‐home policies on systems engineers and the general population

A Cross-Platform Topic Analysis of the Nazi Narrative on Twitter and Telegram during the 2022 Russian Invasion of Ukraine

Pheonix at SemEval-2020 Task 5: Masking the Labels Lubricates Models for Sequence Labeling

Contact Info

Product

Resources

About