Revealing a Country’s Government Discourse Through BERT-based Topic Modeling in the US Presidential Speeches DOI

Rafi Mahmud Zain,

Sajarwo Anggai,

Tukiyat Tukiyat

и другие.

Опубликована: Окт. 9, 2024

Язык: Английский

Two Decades of Rheumatology Research (2000-2023): A Dynamic Topic Modeling Perspective DOI Creative Commons
Alfredo Madrid-García, D. Freites, Luis Rodríguez‐Rodríguez

и другие.

medRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Июнь 9, 2024

Abstract Background Rheumatology has experience notably changes in last decades. New drugs, including biologic agents and janus kinase inhibitors, have bloosom. Concepts such as window of opportunity , arthralgia suspicious for progression or difficult-to-treat rheumatoid arthritis appeared; new management approaches strategies treat-to-target become popular. Statistical learning methods, gene therapy, telemedicine precision medicine are other advancements that gained relevance the field. To better characterise research landscape advances rheumatology, automatic efficient based on natural language processing should be used. The objective this study is to use topic modeling techniques uncover key topics trends rheumatology conducted 23 years. Methods This analysed 96,004 abstracts published between 2000 December 31, 2023, drawn from 34 specialised journals obtained PubMed. BERTopic, a novel approach considers semantic relationships among words their context, was used topics. Up 30 different models were trained. Based number topics, outliers coherence score, two them finally selected, manually labeled by rheumatologists. Word clouds hierarchical clustering visualizations computed. Finally, hot cold identified using linear regression models. Results Abstracts classified into 45 47 most frequent arthritis, systemic lupus erythematosus osteoarthritis. Expected COVID-19 JAK inhibitors after conducting dynamic modeling. Topics spinal surgery bone fractures years, however, antiphospholipid syndrome, septic lost momentum. Conclusions Our utilized advanced analyse landscape, identify themes emerging trends. results highlight varied nature research, illustrating how interest certain shifted over time.

Язык: Английский

Процитировано

2

Revealing a Country’s Government Discourse Through BERT-based Topic Modeling in the US Presidential Speeches DOI

Rafi Mahmud Zain,

Sajarwo Anggai,

Tukiyat Tukiyat

и другие.

Опубликована: Окт. 9, 2024

Язык: Английский

Процитировано

0