A Temporal Topic Model for Noisy Mediums

Churchill, Rob; Singh, Lisa; Kirov, Christo

doi:10.1007/978-3-319-93037-4_4

Cited by 15 publications

(11 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Like PMI, diversity does not require knowledge of ground truth topics. Methods similar to topic diversity have appeared under diferent names over the years, including topic uniqueness [53] and topic overlap [21].…”

Section: Coherencementioning

confidence: 99%

The Evolution of Topic Modeling

2022

Self Cite

View full text Add to dashboard Cite

Topic models have been applied to everything from books to newspapers to social media posts in an effort to identify the most prevalent themes of a text corpus. We provide an in-depth analysis of unsupervised topic models from their inception to today. We trace the origins of different types of contemporary topic models, beginning in the 1990s, and we compare their proposed algorithms, as well as their different evaluation approaches. Throughout, we also describe settings in which topic models have worked well and areas where new research is needed, setting the stage for the next generation of topic models.

show abstract

Section: Coherencementioning

confidence: 99%

The Evolution of Topic Modeling

2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, Surian and colleagues uses both topic modeling (LDA) and community detection to cluster opinions about human papillomavirus (HPV) vaccines on Twitter (Surian et al, 2016). Graph-based topic models have also had success, particularly for temporal topic modeling (Churchill et al, 2018;Cataldi et al, 2010;Churchill & Singh, 2020). These models build networks or graphs, where each node represents a word or phrase in the document collection and edges exist between nodes that cooccur within the same post.…”

Section: Descriptive Modelsmentioning

confidence: 99%

“…Here each group represents a different topic. In the context of social media data, for example, graphs have been used to build topics for the 2016 US presidential election (Churchill et al, 2018;Churchill & Singh 2020).…”

Section: Descriptive Modelsmentioning

confidence: 99%

Modeling Considerations for Quantitative Social Science Research Using Social Media Data

Ladd¹,

Ryan²,

Singh³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, the fourth in a series of white papers, we provide a summary of the discussions and future directions that came from the topical meeting that focused on model construction with social media data. A particularly interesting aspect of this meeting was, in our view, discussion of the different disciplines’ requirements and approaches to modeling and the different considerations that are used to assess model fit.

show abstract

“…Over the last two years, much work has focused on extracting these types of signals from newspaper and public social media data, including detecting relevant dynamic topics and events from newspapers [2,11,21,31,44,45]. We take advantage of this previous work to extract topic buzz, sentiment (as a proxy for perception), and event volume as an initial set of social media and newspaper related signals.…”

Section: Events Buzz and Perceptionmentioning

confidence: 99%

Blending Noisy Social Media Signals with Traditional Movement Variables to Predict Forced Migration

Singh

Wahedi

Wang

et al. 2019

Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

Worldwide displacement due to war and conflict is at all-time high. Unfortunately, determining if, when, and where people will move is a complex problem. This paper proposes integrating both publicly available organic data from social media and newspapers with more traditional indicators of forced migration to determine when and where people will move. We combine movement and organic variables with spatial and temporal variation within different Bayesian models and show the viability of our method using a case study involving displacement in Iraq. Our analysis shows that incorporating open-source generated conversation and event variables maintains or improves predictive accuracy over traditional variables alone. This work is an important step toward understanding how to leverage organic big data for societal-scale problems. CCS CONCEPTS • Information systems → Data mining; • Human-centered computing → Social engineering (social sciences);

show abstract

A Temporal Topic Model for Noisy Mediums

Cited by 15 publications

References 10 publications

The Evolution of Topic Modeling

The Evolution of Topic Modeling

Modeling Considerations for Quantitative Social Science Research Using Social Media Data

Blending Noisy Social Media Signals with Traditional Movement Variables to Predict Forced Migration

Contact Info

Product

Resources

About