A Computational Approach to Politeness with Application to Social Factors

TL;DR

This paper introduces a computational framework for detecting politeness in language, using a new annotated corpus and a classifier that links politeness markers to social factors across domains.

Contribution

It presents a novel corpus, evaluates politeness theory, and develops a domain-independent classifier that links politeness to social power and gender.

Findings

Politeness correlates with social power in Wikipedia and Stack Exchange.

The classifier achieves near-human accuracy across domains.

Politeness varies by gender and community in preliminary analysis.

Abstract

We propose a computational framework for identifying linguistic aspects of politeness. Our starting point is a new corpus of requests annotated for politeness, which we use to evaluate aspects of politeness theory and to uncover new interactions between politeness markers and context. These findings guide our construction of a classifier with domain-independent lexical and syntactic features operationalizing key components of politeness theory, such as indirection, deference, impersonalization and modality. Our classifier achieves close to human performance and is effective across domains. We use our framework to study the relationship between politeness and social power, showing that polite Wikipedia editors are more likely to achieve high status through elections, but, once elevated, they become less polite. We see a similar negative correlation between politeness and power on Stack Exchange, where users at the top of the reputation scale are less polite than those at the bottom. Finally, we apply our classifier to a preliminary analysis of politeness variation by gender and community.