Paper/Danescu-Niculescu-Mizil2013
author: - Cristian Danescu-Niculescu-Mizil - Robert West - Dan Jurafsky - Jure Leskovec - Christopher Potts url: “http://dl.acm.org/citation.cfm?id=2488416”
No country for old members: user lifecycle and linguistic change in online communities
Questions
Much sociolinguistic research has relied on the adult language stability assumption: under the critical assumption that individuals’ speech patterns are largely fixed by early adulthood, older speakers’ language can be employed as a proxy for the linguistic state of the community at an earlier stage.
Can we understand the linguistic changes in online communities in terms of “age” in the community?
Data and methods
- Consider monthly snapshosts to look at the language usage.
- bigram language models with Katz back-off smoothing (Katz’s back off model)