Detecting and ordering adjectival scalemates

TL;DR

This paper introduces a pattern-based approach to automatically identify and order adjectival scalemates from text corpora, improving the accuracy of scale inference through lexical patterns and similarity measures.

Contribution

It presents a novel pattern-based method for inferring and ordering adjectival scales, with an enhanced filtering process and comprehensive evaluation against existing techniques.

Findings

The method effectively infers adjectival scales from corpora.

It outperforms or matches current state-of-the-art in scale inference.

The approach is validated on standard and new evaluation sets.

Abstract

This paper presents a pattern-based method that can be used to infer adjectival scales, such as <lukewarm, warm, hot>, from a corpus. Specifically, the proposed method uses lexical patterns to automatically identify and order pairs of scalemates, followed by a filtering phase in which unrelated pairs are discarded. For the filtering phase, several different similarity measures are implemented and compared. The model presented in this paper is evaluated using the current standard, along with a novel evaluation set, and shown to be at least as good as the current state-of-the-art.