From Incremental Meaning to Semantic Unit (phrase by phrase)

TL;DR

This paper presents an experimental method for detecting minimal semantic units in text using word embeddings and parser-based features, focusing on incremental compositional embeddings for multiword expressions.

Contribution

It introduces a novel incremental approach combining embeddings and parsing features for semantic unit detection within the DiMSUM task.

Findings

Effective detection of minimal semantic units

Improved performance with combined embeddings and parser features

Demonstrated applicability in SemEval 2016 Task 10

Abstract

This paper describes an experimental approach to Detection of Minimal Semantic Units and their Meaning (DiMSUM), explored within the framework of SemEval 2016 Task 10. The approach is primarily based on a combination of word embeddings and parserbased features, and employs unidirectional incremental computation of compositional embeddings for multiword expressions.