From: Smart literature review: a practical topic modelling approach to exploratory literature review
Reference | Data | Method | Intended use | Size |
---|---|---|---|---|
DiMaggio et al. [12] | Newspapers | LDA | Identify central concepts in news coverage | 8000 |
Grimmer [17] | Press releases | Own developed method | Development of a Bayesian topic model | 24,000 |
Quinn et al. [30] | Speeches (Text) | Own developed method | Development of a statistical learning model | 118,000 |
Jockers and Mimno [21] | Books | LDA | Identify literature themes | 3346 |
Baum, [5] | Speeches (Video) | LDA | Identify topics of German politicians | 2581 |
Ghosh and Guha [16] | Tweets | LDA | Identify tweets related to obesity | 2,581,283 |
Evans [15] | Newspapers | LDA | Identify subjects of discussion | 14,952 |
Guo et al. [19] | Tweets | Dictionary-based analysis and LDA | Compare dictionary-based analysis vs. LDA | 77,000,000 |
Jacobi et al. [20] | Newspapers | LDA | Show the usefulness of LDA | Newspaper articles from 1945 to 2013 |
Maier et al. [27] | Web pages | LDA | Investigate the validity and reliability of LDA | 344,456 |
Bonilla and Grimmer [9] | Newspapers | LDA | Investigate impact of terror alerts in US media | 50,000 |
Elgesem et al. [13] | Blogs | LDA | Identify topics in blogs regarding the arrest of Edward Snowden | 15,000 |
Elgesem et al. [14] | Blogs | LDA | Investigate how climate change is discussed in blogs | 1,300,000 |
Koltsova and Koltcov [24] | Web forum posts | LDA | Investigate the political agenda of Russians in LiveJournal | > 100,000 |
Welbers et al. [4] | Newspapers | LDA | Validate the use of LDA | 99,572 |
Parra et al. [29] | Tweets | LDA | Investigate how Twitter has been used in academic conferences | 109,076 |