Dataset Groups Activity Stream DIR dataset The DIR dataset is a semi-synthetic dataset constructed from the matrix X, whose columns are from a Dirichlet distribution with parameters (0.05, 0.05,..., 0.05). BibTex: @dataset{Yuanzhi_Li_and_Yingyu_Liang_2024, abstract = {The DIR dataset is a semi-synthetic dataset constructed from the matrix X, whose columns are from a Dirichlet distribution with parameters (0.05, 0.05,..., 0.05).}, author = {Yuanzhi Li and Yingyu Liang}, doi = {10.57702/x9havkcx}, institution = {No Organization}, keyword = {'Dirichlet Distribution', 'Semi-synthetic Dataset', 'Topic Modeling'}, month = {dec}, publisher = {TIB}, title = {DIR dataset}, url = {https://service.tib.eu/ldmservice/dataset/dir-dataset}, year = {2024} }