Topic Modeling of Conflict Ad Hoc Discussions in Social Networks

Blekanov, Ivan S.; Tarasov, Nikita; Maksimov, Alexey

doi:10.1145/3274856.3274881

Cited by 10 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unsupervised topic modeling strategies, such as BTM, are methods particularly well suited for sorting short text (such as the 280-character limit for tweets) into highly prevalent themes without the need for predetermined coding or a training/labelled dataset to classify specific content. This is particularly useful in characterizing large volumes of unstructured data where predefined themes are unavailable, such as in the case of emerging social movements, novel disease outbreaks, and other emergency events where information changes rapidly [ 10 , 56 , 60 , 61 , 63 , 64 , 76 , 77 ]. The corpus of tweets containing the 5G keywords was categorized into highly correlated topic clusters using BTM based on splitting all text into a bag of words and then producing a discrete probability distribution for all words for each theme that places a larger weight on words that are most representative of a given theme [ 78 ].…”

Section: Methodsmentioning

confidence: 99%

“…disease outbreaks, and other emergency events where information changes rapidly [10,56,60,61,63,64,76,77]. The corpus of tweets containing the 5G keywords was categorized into highly correlated topic clusters using BTM based on splitting all text into a bag of words and then producing a discrete probability distribution for all words for each theme that places a larger weight on words that are most representative of a given theme [78].…”

Section: Plos Onementioning

confidence: 99%

See 1 more Smart Citation

Detecting nuance in conspiracy discourse: Advancing methods in infodemiology and communication science with machine learning and qualitative content coding

Haupt,

Chiu,

Chang

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

The spread of misinformation and conspiracies has been an ongoing issue since the early stages of the internet era, resulting in the emergence of the field of infodemiology (i.e., information epidemiology), which investigates the transmission of health-related information. Due to the high volume of online misinformation in recent years, there is a need to continue advancing methodologies in order to effectively identify narratives and themes. While machine learning models can be used to detect misinformation and conspiracies, these models are limited in their generalizability to other datasets and misinformation phenomenon, and are often unable to detect implicit meanings in text that require contextual knowledge. To rapidly detect evolving conspiracist narratives within high volume online discourse while identifying nuanced themes requiring the comprehension of subtext, this study describes a hybrid methodology that combines natural language processing (i.e., topic modeling and sentiment analysis) with qualitative content coding approaches to characterize conspiracy discourse related to 5G wireless technology and COVID-19 on Twitter (currently known as ‘X’). Discourse that focused on correcting 5G conspiracies was also analyzed for comparison. Sentiment analysis shows that conspiracy-related discourse was more likely to use language that was analytic, combative, past-oriented, referenced social status, and expressed negative emotions. Corrections discourse was more likely to use words reflecting cognitive processes, prosocial relations, health-related consequences, and future-oriented language. Inductive coding characterized conspiracist narratives related to global elites, anti-vax sentiment, medical authorities, religious figures, and false correlations between technology advancements and disease outbreaks. Further, the corrections discourse did not address many of the narratives prevalent in conspiracy conversations. This paper aims to further bridge the gap between computational and qualitative methodologies by demonstrating how both approaches can be used in tandem to emphasize the positive aspects of each methodology while minimizing their respective drawbacks.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Plos Onementioning

confidence: 99%

Detecting nuance in conspiracy discourse: Advancing methods in infodemiology and communication science with machine learning and qualitative content coding

Haupt,

Chiu,

Chang

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Research by [11] concluded that the BTM method is the best and most stable based on all coherence measures, whereas the non-specific short text LDA model showed not suitable without additional data pre-processing mainly due to the data availability. Research by [12] concluded that BTM is somewhat better than LDA and word network topic model but worse than the topic keyword model for any classification cases except for the book reviews data set, which is characterized by relatively short documents that seem to be better handled by BTM.…”

Section: Literature Reviewmentioning

confidence: 99%

Public Hospital Review on Map Service with Part of Speech Tagging and Biterm Topic Modeling

Mak’ruf¹,

Bramantoro²

2023

IJCDS

View full text Add to dashboard Cite

Health is a basic human need. Increasing standard of life influences the quality of health. It demands health service providers improve quality better service and provide satisfaction for consumers as health care users. Hospital as one of the means of health services is required to improve the quality of services. The study found that customers tend to choose private hospitals over public hospitals because of service factors. Public hospitals are gradually being abandoned. This is challenging for public hospitals to improve services. One of the scientific techniques to know the level of public satisfaction with health services, especially hospitals, is to examine the reviews from users. In this research, we use public mapping review as a data source. From the review, it can be known what topics are mostly discussed specifically for various ratings. A suitable model is required to find out the topics in the review that have a low rating so that it can be used as a suggestion for improving health services. Topic modeling is best achieved through the text mining method. The study proposed the use biterm topic model suitable for short text in the reviews of the map platform. Short text reviews are characterized by sparse data, the small number of words that appear as topic builders, and rare word contexts related to the topic. The result shows that biterm topic model can produce topics with an acceptable combination of accuracy and performance. The model with the addition of part-of-speech tagging on noun-only is unable to increase the accuracy of the model, compared to the one without the addition of part-of-speech tag. However, the addition of part-of-speech tagging on noun-only can improve the performance of the model.

show abstract

“…The works by Bodrunova and colleagues appear to be the only continuous effort (since 2013) to combine topic modeling for Twitter with various other instruments of automated text analysis, also in comparison with other languages (Bodrunova et al 2019a, c). Thus, we have tested three topic modeling algorithms, namely unsupervised LDA, WNTM, and BTM (Blekanov et al 2018), and have shown that BTM works best, as measured by normalized PMI and Umass (see below). We have also applied BTM to detect the dynamics of topicality in conflictual discussions (Smoliarova et al 2018) and have demonstrated that the saliency of topics in time may help detect pivotal points in mediated discussions.…”

Section: Topic Modeling For the Russian Twittermentioning

confidence: 99%

Topic Modeling in Russia: Current Approaches and Issues in Methodology

Bodrunova

2020

The Palgrave Handbook of Digital Russia Studies

View full text Add to dashboard Cite

Topic modeling as an instrument of probabilistic clustering for text collections has gained particular attention within the computational social science in Russia. This chapter looks at how topic modeling techniques have been developed and employed by the Russian scholars, both for Russian and other languages. We divide the works on topic modeling into methodological, applied, relational, and those dedicated to modeling quality assessment. While the methodological studies are the most developed, the works explaining the substance of the Russian-language discussions cover an important niche in political and social science. However, there is a gap between method-oriented works that treat Russian as “language as such” and the works by computational linguists who focus on Russian but treat topic modeling as a method of secondary importance.

show abstract

Topic Modeling of Conflict Ad Hoc Discussions in Social Networks

Cited by 10 publications

References 3 publications

Detecting nuance in conspiracy discourse: Advancing methods in infodemiology and communication science with machine learning and qualitative content coding

Detecting nuance in conspiracy discourse: Advancing methods in infodemiology and communication science with machine learning and qualitative content coding

Public Hospital Review on Map Service with Part of Speech Tagging and Biterm Topic Modeling

Topic Modeling in Russia: Current Approaches and Issues in Methodology

Contact Info

Product

Resources

About