Hi, I’m Naitian.
That’s pronounced naɪtjen like 🌙 💴.
I’m a PhD student at the UC Berkeley School of Information, where I’m advised by David Bamman and supported by the NSF graduate research fellowship.
My research involves applying large-scale data analysis and deep learning methods to studying culture, often through a variationist sociolinguistic analytic framework. This spans the fields of NLP, computational social science, and cultural analytics. I also care a lot about the news, data journalism, data visualization and crossword puzzles.
If you are interested in doing research or grad school, I am always happy to chat about my experiences. If you are a Berkeley undergrad interested in doing cultural analytics research, you should apply for a UROP opportunity with my advisor, David Bamman.
Updates
- Apr. 2024: Our paper on linguistic variation in memes has been accepted! (whoo!)
- Oct. 2023: I presented at NWAV51 about how zero-shot text-to-speech models erase phonological variation (schwa!)
- Mar. 2023: The NSF awarded me the Graduate Research Fellowship (wild!)
- Aug. 2022: I started my PhD at UC Berkeley (omg!)
- Jun. 2022: I ran my first marathon (oof!)
- Mar. 5, 2022: Ran my first half marathon race (whoa!)
- Sometime between Dec. 2021 and Jan. 2022: Graduated from the University of Michigan (wow!)
- Dec. 22, 2021: Received an Honorable Mention for the CRA Outstanding Undergraduate Researcher Award (neat!)
- Nov. 26, 2021: Launched the Spalling Bie (cool!)
Publications
Much work in the space of NLP has used computational methods to explore sociolinguistic variation in text. In this paper, we argue that memes, as multimodal forms of language comprised of visual templates and text, also exhibit meaningful social variation.
[Website] [PDF]
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
In times of distress, we frequently go online to seek social support and condolence. But effectively providing that support to others is easier said than done. This study aims to computationally identify mechanisms and strategies for delivering effective and impactful condolence on social media.
[Website] [PDF]
Code
I am writing or have written code for: The Michigan
Daily as the managing online
editor, NBC News as a Data
Graphics intern, the Michigan Data Science
Team as a project leader, Capital One as a software engineering intern
(x2 summers) and, of course, myself as naitian
.
Here is a sampler of my work.
The Spalling Bie is just like the New York Times Spelling Bee, except you only get points for fake words that sound plausible. [Link]
I did research, data collection, analysis and graphics for an NBC News story tracking where Americans could expect to find pharmacies that would carry the Covid-19 vaccine. [Link]
As part of a challenge, I used a database of book titles from Amazon and a part-of-speech tagger to construct grammatically correct sentences using only book titles. Earned an honorable mention from Randall Munroe, creator of XKCD. [Link]
A different kind of New Year’s countdown, IYSP takes inspiration from the internet memes about “starting the year off right.” [Link]