Does ‘well-being’ translate on Twitter?

By Laura K Smith, Salvatore Giorgi, Rishi Solanki, Johannes C. Eichstaedt, H. A. Schwartz, Muhammad Abdul-Mageed, Anneke Buffone, Lyle Ungar. 2016

12 citations recorded by the source corpus.

We investigate whether psychological well-being translates across English and Spanish Twitter, by building and comparing source language and automatically translated weighted lexica in English and Spanish. We find that the source language models perform substantially better than the machine translated versions. Moreover, manually correcting translation errors does not improve model performance, suggesting that meaningful cultural information is being lost in translation. Further work is needed to clarify when automatic translation of well-being lexica is effective and how it can be improved for cross-cultural analysis.

Subfields: Artificial Intelligence, Social Psychology, Sociology and Political Science, Sentiment & Opinion Analysis

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

1. Graham Neubig
2. Yulia Tsvetkov
3. Orhan Firat
4. Sebastian Ruder
5. Melvin Johnson
6. Mohit Bansal
7. Ivan Vulic
8. Luke S. Zettlemoyer

In Social Psychology

1. Yulia Tsvetkov
2. Dan Jurafsky
3. David Jurgens

In Sociology and Political Science

1. Yulia Tsvetkov
2. Dan Jurafsky
3. David Jurgens
4. Chan Young Park
5. Maarten Sap

In Sentiment & Opinion Analysis

1. Dan Jurafsky
2. Yulia Tsvetkov

Citations in the corpus

• XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization (2020, background): “For instance, it does not capture differences in classification performance that are due to cultural differences (Mohammad et al., 2016; Smith et al., 2016).”
• Incorporating Dialectal Variability for Socially Equitable Language Identification (2017, methodology): “Setup A list of health-related terms was compiled from lexicons for influenza (Lamb et al., 2013); psychological well-being (Smith et al., 2016; Preot¸iuc-Pietro et al., 2015); and temporal orientation lexica correlated with age, gender and personality traits (Park et al., 2016).”
• Ranking Transfer Languages with Pragmatically-Motivated Features for Multilingual Sentiment Analysis (2020, background): “…the cross-lingual task is driven by pragmatic knowledge, as in the cross-lingual analysis of sentiment and emotion.1 Expression of subtle sentiment and emotion, such as subjective wellbeing (Smith et al., 2016), anger (Oster, 2019), or irony (Karoui et al., 2017), varies significantly by culture.”