NEWS

---

seededlda 1.4.4

• Fix texts to pass checks on CRAN.

seededlda 1.4.3 (2025-09-28)

• Add as.dictionary to create a dictionary object from topic terms.
• Suppress messages from internal functions.
• Move quanteda from Depends to Imports.

seededlda 1.4.2 (2025-01-07)

• Fix tests for quanteda v4.2.0.

seededlda 1.4.1 (2024-09-06)

• Fix regression in 1.4.0 on Linux-like OS.

seededlda 1.4.0 (2024-09-05)

• Use configure to link the TBB library on MacOS.
• Add adjust_alpha as an experimental argument to optimize alpha automatically.
• Add update_model to update terms of existing models to classify documents with unseen words more accurately.

seededlda 1.3.2 (2024-07-03)

• Improve the way to convert std::vector to arma::mat.

seededlda 1.3.1 (2024-06-25)

• Fix C++ code for Armadillo v14.

seededlda 1.3.0 (2024-06-19)

• Add perplexity() to compute perplexity scores of fitted LDA models.
• Improve documentation.

seededlda 1.2.1 (2024-04-11)

• Fix tests on systems when the TBB library is unavailable.

seededlda 1.2.0 (2024-04-10)

• The RcppParallel package is no longer required as the TBB library in the operating system (Linux and MacOS) or Rtools (Windows) is used.
• Linux and MacOS must have the TBB library to enable parallel computing before installing this package from the source.

seededlda 1.1.1

• Allow alpha and beta to be a vector for asymmetric Dirichlet priors.

seededlda 1.1.0 (2023-07-01)

• Remove uniform to simplify the computation of seed word weights.
• Add levels argument to better handle hierarchical dictionaries.

seededlda 1.0.1 (2023-06-12)

• Fix the error when textmodel_seqlda() is called.
• Save values in the Array object in double to avoid rounding error (#60).

seededlda 1.0.0 (2023-05-31)

• Add auto_iter to textmodel_seededlda() and textmodel_lda() to stop Gibbs sampling automatically before max_iter is reached.
• Add batch_size to textmodel_seededlda() and textmodel_lda() to enable the distributed LDA algorithm for parallel computing.

seededlda 0.9.0 (2023-04-30)

• Add the gamma parameter to textmodel_seededlda() and textmodel_lda() for sequential classification.
• Add textmodel_seqlda() as as short cut for textmodel_lda(gamma = 0.5).
• Improve the calculation of weights for seed words.
• Add the regularize argument to divergence() for the regularized topic divergence measure.

seededlda 0.8.4 (2023-03-23)

• Fix for deprecation in Matrix 1.5-4.

seededlda 0.8.3 (2023-03-17)

• Add data_corpus_moviereviews to the package to reduce dependency.

seededlda 0.8.2 (2022-10-09)

• Add min_prob and select to topics() for greater flexibility
• Change the divergence measure from Kullback-Leibler to Jensen-Shannon.
• Add weighted, min_size, select to divergence() for regularized topic divergence scores.

seededlda 0.8.1 (2022-03-28)

• Change textmodel_seededlda() to set positive integer values to residual.
• Fix a bug in textmodel_seededlda() that ignores n-grams when concatenator is not "_".
• Change topics() to return document names.
• Add divergence() to optimize the number of topics or the seed words (#26).

seededlda 0.8.0 (2022-01-07)

• Add the model argument to textmodel_lda() to replace predict().

seededlda 0.7.0

• Change the textmodel_seededlda object to save dictionary and related settings (#18)

seededlda 0.6.0 (2021-04-08)

• Add predict() to identify topics of unseen documents (#9)
• Allow selecting seed words based on their frequencies using dfm_trim() in textmodel_seededlda() via ... (#8)

seededlda 0.5.1 (2020-12-17)

• Change topics() to return factor with NA for empty documents
• Fix a bug in initializing LDA that leads to incorrect phi (#4 and #6)

seededlda 0.5 (2020-09-10)

• Implement original LDA estimator using the LDAGibbs++ library