Searching for PETs: Using Distributional and Sentiment-Based Methods to Find Potentially Euphemistic Terms

TL;DR

This paper introduces a linguistically driven method combining distributional similarity and sentiment analysis to identify potentially euphemistic terms across various sensitive topics, demonstrating promising results in detecting PETs.

Contribution

It presents a novel approach that integrates distributional and sentiment-based techniques for detecting euphemistic language, advancing methods for linguistic analysis of sensitive content.

Findings

Effective detection of single and multi-word PETs

Demonstrated approach's efficacy on euphemism corpus

Potential for sentiment-based methods in euphemism detection

Abstract

This paper presents a linguistically driven proof of concept for finding potentially euphemistic terms, or PETs. Acknowledging that PETs tend to be commonly used expressions for a certain range of sensitive topics, we make use of distributional similarities to select and filter phrase candidates from a sentence and rank them using a set of simple sentiment-based metrics. We present the results of our approach tested on a corpus of sentences containing euphemisms, demonstrating its efficacy for detecting single and multi-word PETs from a broad range of topics. We also discuss future potential for sentiment-based methods on this task.