Papers with discrete and continuous representations
Topic or Style? Exploring the Most Useful Features for Authorship Attribution (C18-1)
Copied to clipboard
| Challenge: | Existing approaches to authorship attribution rely on individual's writing style and/or preferred topics. |
| Approach: | They analyse four widely used datasets to explore how different types of features affect authorship attribution accuracy under varying conditions. |
| Outcome: | The proposed model outperforms the state-of-the-art on two out of the four datasets used. |
Adaptive Mixed Component LDA for Low Resource Topic Modeling (2021.eacl-main)
Copied to clipboard
| Challenge: | Probabilistic topic models in low resource settings are faced with less reliable estimates due to sparsity of discrete word co-occurrence counts. |
| Approach: | They propose a mixture model which interpolates between discrete and continuous topic-word distributions and utilises pre-trained embeddings to improve topic coherence. |
| Outcome: | The proposed model outperforms fully discrete, fully continuous, and static mixture models on topic coherence in low resource settings. |