Using socially-aware word embeddings to infer variation in the meaning of political terms over time with Hubert AU, DPhil student at University of Oxford
Semantic change occurs organically as languages and current events evolve. For example, “Leave” and “Remain” were understood differently before and after the Brexit campaign and vote.
|An Institute for Data Science and Artificial Intelligence seminar|
|Date||17 May 2023|
|Time||14:00 to 15:00|
|Place||Streatham Court Old C |
To delivered hybrid using Zoom
This talk explores the topic of semantic change and its significance, particularly in the context of social networks. We demonstrate a new computational method for detecting lexical semantic change by leveraging metadata that encodes network relationships between speakers. Our approach is demonstrated using parliamentary data from around the world, including the UK Parliament, which provides a rich source of language use in highly structured social contexts (i.e., political party membership). We also expand beyond the UK Parliament and the Brexit campaign, applying the methods to other events and parliamentary debates in different countries and languages. Our method provides a powerful tool for studying semantic change, which can have important implications for understanding how language and meaning evolve over time.
Delivery and Registration:
To be delivered hybrid. To manage the registrations, we ask participants to complete a simple form, which closes on the morning of the seminar, but please don't let that put you off. If you do miss the registration cut-off, then please email IDSAI. To register, please click here.
If you have any queries, please contact email@example.com.
The seminar forms part of the IDSAI Research Seminar Series for 2022-2023. Click here to find out more.
Streatham Court Old C