Revision as of 09:45, 5 June 2021 edit OpenNotes1 (talk \| contribs) Extended confirmed users 2,834 edits {{Natural Language Processing}} ← Previous edit		Revision as of 10:49, 25 June 2021 edit undo Modulato (talk \| contribs) Extended confirmed users 777 edits →Finding topics: +LDA Next edit →
Line 29: ===Finding topics=== [[Multivariate analysis]] of the document-term matrix can reveal topics/themes of the corpus. Specifically, [[latent semantic analysis]] and [[data clustering]] can be used, and, more recently, [[probabilistic latent semantic analysis]] with its generalization [[Latent Dirichlet allocation]], and [[non-negative matrix factorization]], have been found to perform well for this task. ==See also==

Document-term matrix: Difference between revisions