A computational approach to politeness with application to social factors

By Cristian Danescu-Niculescu-Mizil, Moritz Sudhof, Dan Jurafsky, Jure Leskovec, Christopher Potts. 2013

507 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Language and Linguistics, Literature and Literary Theory, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Language and Linguistics

In Literature and Literary Theory

In Natural Language Processing

Citations in the corpus

RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models (2020, background): “To combat these issues, we encourage further work on detecting and controlling different types of toxicity and undesirable social biases in generation, e.g., rudeness (Danescu-Niculescu-Mizil et al., 2013), hate speech (Golbeck et al., 2017), or microaggressions (Breitfeller et al., 2019).”
Politeness Transfer: A Tag and Generate Approach (2020, methodology): “We circumscribe the scope of politeness for the purpose of this study as follows: First, we adopt the data driven deﬁnition of politeness proposed by (Danescu-Niculescu-Mizil et al., 2013).”
Can Large Language Models Transform Computational Social Science? (2023, methodology): “We evaluate on the Stanford Politeness Corpus (Danescu-Niculescu-Mizil et al., 2013).”
A Just and Comprehensive Strategy for Using NLP to Address Online Abuse (2019, methodology): “As NLP solutions have been developed for other social dimensions of language such as politeness (Danescu-Niculescu-Mizil et al., 2013; Munkova et al., 2013; Chhaya et al., 2018) and formality (Brooke et al., 2010; Sheikha and Inkpen, 2011; Pavlick and Tetreault, 2016), these methods could be…”
Beyond Distributive Fairness in Algorithmic Decision Making: Feature Selection for Procedurally Fair Learning (2018, background): “1Previous work has demonstrated the importance and validity of measuring perception of normative phenomena via human surveys, for example language impartiality (Zafar, Gummadi, and DanescuNiculescu-Mizil 2016) and politeness (Danescu-Niculescu-Mizil et al. 2013).”
Predicting In-Game Actions from Interviews of NBA Players (2019, background): “In the emerging ﬁeld of computational social science, there is a substantial effort to harness linguistic signals to better answer scientiﬁc questions (Danescu-Niculescu-Mizil et al. 2013).”
A Bad Workman Blames His Tweets? The Consequences of Citizens’ Uncivil Twitter Use When Interacting with Party Candidates (2016, background): “For example, Danescu-Niculescu-Mizil et al. (2013) report a maximum of 84% accuracy in coding of impoliteness in conversations on Wikipedia, only slightly below 86% agreement using human coding. probability of being classified as engaging are receiving more responses by ordinary citizens, under the…”
Does It Capture STEL? A Modular, Similarity-based Linguistic Style Evaluation Framework (2021, methodology): “Binary and more fine-grained style classification has been employed on word, text fragment as well as document level (Danescu-Niculescu-Mizil et al., 2013; Pavlick and Nenkova, 2015; Preoţiuc- Pietro et al., 2016; Pavlick and Tetreault, 2016; Kang et al., 2019).”
The Importance of Modeling Social Factors of Language: Theory and Practice (2021, background): “They go a long way to explaining other socially motivated constructs, such as power imbalances or politeness, which in turn can also be inferred from dialogue (Prabhakaran et al., 2012; Danescu-Niculescu-Mizil et al., 2013a).”
Conversations Gone Awry: Detecting Early Signs of Conversational Failure (2018, methodology): “Recent studies have computationally operationalized prior formulations of politeness by extracting linguistic cues that reflect politeness strategies (Danescu-Niculescu-Mizil et al., 2013; Aubakirova and Bansal, 2016).”
Computational Sociolinguistics: A Survey (2016, methodology): “Review sites, such as TripAdvisor (Michael and Otterbacher 2014), IMDb (Hemphill and Otterbacher 2012) and beer review communities (Danescu-Niculescu-Mizil et al. 2013b), have also been used in studies on language in online communities.”
Language from police body camera footage shows racial disparities in officer respect (2017, background): “morning|good evening|how are you doing|how ’s it going))" [4]”