English

Low-dimensional Embeddings for Interpretable Anchor-based Topic Inference

Computation and Language 2017-11-21 v1

Abstract

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 algorithm often selects poor anchor words, reducing topic quality and interpretability. Rather than finding an approximate convex 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.

Keywords

Cite

@article{arxiv.1711.06826,
  title  = {Low-dimensional Embeddings for Interpretable Anchor-based Topic Inference},
  author = {Moontae Lee and David Mimno},
  journal= {arXiv preprint arXiv:1711.06826},
  year   = {2017}
}
R2 v1 2026-06-22T22:50:13.063Z