You Write like You Eat: Stylistic Variation as a Predictor of Social Stratification

Basile, Angelo; Gatt, Albert; Nissim, Malvina

doi:10.18653/v1/p19-1246

Cited by 4 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The averaged results were reported. Results For the challenging task of SES prediction, all models attain moderate performance that is consistently above chance level (Table 13), echoing previous findings (Flekova et al, 2016;Basile et al, 2019). Compared to the fine-tuned RoBERTa, the idiolectal features have filtered out some SES-related variations, which could be related to domain-specific information.…”

Section: F Additional Analysis: Characterizing Sociolectssupporting

confidence: 79%

“…Dataset compilation From the test set, we created a small subset of high socioeconomic status (SES) users and low SES users by using the prices of the reviewed products as a proxy. We verified that there is a clear distinction in readability between high SES and low SES groups, which is a reliable linguistic indicator of SES (Flekova et al, 2016;Basile et al, 2019).…”

Section: F Additional Analysis: Characterizing Sociolectssupporting

confidence: 62%

“…The distribution of product domains is displayed in Figure 5. Label validation After dataset creation, we verified that there was linguistic variation conditioning on socio-economic status in the dataset by measuring several readability metrics, linguistic indicators shown to correlate reliably with socio-economic status (Flekova et al, 2016;Basile et al, 2019).…”

Section: F Additional Analysis: Characterizing Sociolectsmentioning

confidence: 98%

“…If a user consistently reviews expensive products, it is more likely that this user is associated with high socioeconomic status. This method has been used in a study that surveyed socio-economically related variations (Basile et al, 2019).…”

Section: F Additional Analysis: Characterizing Sociolectsmentioning

confidence: 99%

See 3 more Smart Citations

Idiosyncratic but not Arbitrary: Learning Idiolects in Online Registers Reveals Distinctive yet Consistent Individual Styles

Zhu¹,

Jurgens²

2021

Preprint

View full text Add to dashboard Cite

An individual's variation in writing style is often a function of both social and personal attributes. While structured social variation has been extensively studied, e.g., gender based variation, far less is known about how to characterize individual styles due to their idiosyncratic nature. We introduce a new approach to studying idiolects through a massive crossauthor comparison to identify and encode stylistic features. The neural model achieves strong performance at authorship identification on short texts and through an analogybased probing task, showing that the learned representations exhibit surprising regularities that encode qualitative and quantitative shifts of idiolectal styles. Through text perturbation, we quantify the relative contributions of different linguistic elements to idiolectal variation. Furthermore, we provide a description of idiolects through measuring inter-and intraauthor variation, showing that variation in idiolects is often distinctive yet consistent.

show abstract

Section: F Additional Analysis: Characterizing Sociolectssupporting

confidence: 79%

Section: F Additional Analysis: Characterizing Sociolectssupporting

confidence: 62%

Section: F Additional Analysis: Characterizing Sociolectsmentioning

confidence: 98%

Section: F Additional Analysis: Characterizing Sociolectsmentioning

confidence: 99%

See 2 more Smart Citations

Idiosyncratic but not Arbitrary: Learning Idiolects in Online Registers Reveals Distinctive yet Consistent Individual Styles

Zhu¹,

Jurgens²

2021

Preprint

View full text Add to dashboard Cite

show abstract

A Domain Knowledge Transformer Model for Occupation Profiling

Ai,

Yijia,

Mingyu

2023

Int J Comput Intell Syst

View full text Add to dashboard Cite

Occupation profiling is a subtask of authorship profiling that is broadly defined as an analysis of individuals’ writing styles. Although the problem has been widely explored, no previous studies have attempted to identify Chinese classical poetry. Inspired by Trudgill’s seminal work on stylistic variation as a function of occupation, we present a novel Domain-Knowledge Transformer model to predict a poet’s occupation through their poems’ writing styles. Different from other Indo-European languages, Chinese has rarely used characters and two types of writing forms: traditional Chinese and simplified Chinese. To tackle these problems, we use the language-related component to standardize our input. We also use alphabetization to satisfy the restrictions on rhyming rules and tonal styles. As a special literal form, traditional domain knowledge, for example, named entities, themes, ages and the official career path, is valuable for poet occupation profiling. However, due to the lack of appropriate annotation datasets, it is difficult to recognize these features. Therefore, we proposed the domain knowledge component employing the latent Dirichletal location model to capture the extra theme information and establish named entity dictionaries to recognize the named entity of the datasets in this study. Finally, in the deep learning component, we combine Transformer with a convolutional neural network (CNN) model to perform occupation profiling. The experimental results suggest that our model is effective in this task. Moreover, the results demonstrate an account of other social attribution features of poetry style that are predictive of occupation in this domain.

show abstract