Export Reviews, Discussions, Author Feedback and Meta-Reviews

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. This paper proposes a version of labeled LDA that (1) supports multi-labeled data; (2) produces a label hierarchy. Comparisons with baselines on congressional bills shows improvements in perplexity as well as classification accuracy. The approach is a natural way to handle multi-labeled data, and the hierarchy inference approach appears sound. The paper is generally well-written, and the baseline comparisons are sufficient.