# On the Compositionality Prediction of Noun Phrases using Poincar\'e Embeddings

## TL;DR

This paper introduces a novel method combining hierarchical Poincaré embeddings with distributional semantics to improve the prediction of noun phrase compositionality, achieving significant results over existing models.

## Contribution

The paper presents a new technique that integrates Poincaré embeddings with distributional information for better compositionality prediction, including a supervised approach and publicly released embeddings.

## Key findings

- Significant improvement over state-of-the-art models.
- Effective combination of hierarchical and distributional data.
- Supervised framing yields comparable gains.

## Abstract

The compositionality degree of multiword expressions indicates to what extent the meaning of a phrase can be derived from the meaning of its constituents and their grammatical relations. Prediction of (non)-compositionality is a task that has been frequently addressed with distributional semantic models. We introduce a novel technique to blend hierarchical information with distributional information for predicting compositionality. In particular, we use hypernymy information of the multiword and its constituents encoded in the form of the recently introduced Poincar\'e embeddings in addition to the distributional information to detect compositionality for noun phrases. Using a weighted average of the distributional similarity and a Poincar\'e similarity function, we obtain consistent and substantial, statistically significant improvement across three gold standard datasets over state-of-the-art models based on distributional information only. Unlike traditional approaches that solely use an unsupervised setting, we have also framed the problem as a supervised task, obtaining comparable improvements. Further, we publicly release our Poincar\'e embeddings, which are trained on the output of handcrafted lexical-syntactic patterns on a large corpus.