# Analytical Methods for Interpretable Ultradense Word Embeddings

## TL;DR

This paper explores three rotation-based methods, including a new approach called DensRay, to make ultradense word embeddings interpretable without losing information, and demonstrates their effectiveness in various NLP tasks.

## Contribution

The paper introduces DensRay, a novel, hyperparameter-free rotation method for interpretable word embeddings, and compares it with existing techniques like Densifier and linear SVMs.

## Key findings

- DensRay is computationally efficient and robust.
- All methods improve interpretability in lexicon and analogy tasks.
- Qualitative analysis shows potential for bias removal in embeddings.

## Abstract

Word embeddings are useful for a wide variety of tasks, but they lack interpretability. By rotating word spaces, interpretable dimensions can be identified while preserving the information contained in the embeddings without any loss. In this work, we investigate three methods for making word spaces interpretable by rotation: Densifier (Rothe et al., 2016), linear SVMs and DensRay, a new method we propose. In contrast to Densifier, DensRay can be computed in closed form, is hyperparameter-free and thus more robust than Densifier. We evaluate the three methods on lexicon induction and set-based word analogy. In addition we provide qualitative insights as to how interpretable word spaces can be used for removing gender bias from embeddings.