Paper: Low-dimensional Embeddings for Interpretable Anchor-based Topic Inference

ACL ID D14-1138
Title Low-dimensional Embeddings for Interpretable Anchor-based Topic Inference
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2014
Authors

The anchor words algorithm performs provably efficient topic model inference by finding an approximate convex hull in a high-dimensional word co-occurrence space. However, the existing greedy al- gorithm often selects poor anchor words, reducing topic quality and interpretability. Rather than finding an approximate con- vex hull in a high-dimensional space, we propose to find an exact convex hull in a visualizable 2- or 3-dimensional space. Such low-dimensional embeddings both improve topics and clearly show users why the algorithm selects certain words.