Diachronic degradation of language models: Insights from social media
Natural languages change over time because they evolve to the needs of their users and the socio-technological environment. This study investigates the diachronic accuracy of pre-trained language models for downstream tasks in machine learning and user profiling. It asks the question: given that the social media platform and its users remain the same, how is language changing over time? How can these differences be used to track the changes in the affect around a particular topic? To our knowledge, this is the first study to show that it is possible to measure diachronic semantic drifts within social media and within the span of a few years.
Code (0)
등록된 구현이 없습니다.
Tasks
Word EmbeddingsSimilar Papers 제목 키워드 기반
Contextualized Diachronic Word Representations
Diachronic word embeddings play a key role in capturing interesting patterns about how language evolves over time. Most of the existing work focuses on studying corpora spanning across several decades, which is understan…
Diachronic Word EmbeddingsInductive BiasWord EmbeddingsRevealing COVID-19's Social Dynamics: Diachronic Semantic Analysis of Vaccine and Symptom Discourse on Twitter
Social media is recognized as an important source for deriving insights into public opinion dynamics and social impacts due to the vast textual data generated daily and the 'unconstrained' behavior of people interacting …
Detecting Historical Turning Points in Italian Media: A Complex Systems Approach to a Diachronic News Corpus
The increasing availability of large-scale textual corpora has opened new possibilities for data-driven, quantitative approaches to historical analysis using Natural Language Processing (NLP). However, diachronic corpora…
Language, Place, and Social Media: Geographic Dialect Alignment in New Zealand
This thesis investigates geographic dialect alignment in place-informed social media communities, focussing on New Zealand-related Reddit communities. By integrating qualitative analyses of user perceptions with computat…
Language ModellingUnsupervised detection of diachronic word sense evolution
Most words have several senses and connotations which evolve in time due to semantic shift, so that closely related words may gain different or even opposite meanings over the years. This evolution is very relevant to th…
Time SeriesTime Series AnalysisWord Embeddings