# Computational Linguistics Research | Bruno Gonçalves

Source: https://bgoncalves.com/computational-linguistics/

> Bruno Gonçalves's research in computational linguistics — dialect geography, language change, and the global language network, mapped from large-scale social media and book corpora.

## Computational Linguistics Research

Dialect geography, language change and the global language network, from Twitter and Google Books.

## Overview

Language is one of the richest traces human beings leave behind, and social media made that trace observable at planetary scale for the first time. Bruno Gonçalves's work in computational linguistics uses large corpora of geotagged microblogging messages, digitized books, and translation networks to study how language varies across space, how it changes over time, and how the world's languages connect to one another — across English, Spanish, and Chinese — treating dialect maps and language relationships as measurable, data-driven objects rather than matters of description alone.

## Research Areas

- **Dialect Geography**

  Mapping lexical variation at global scale from geotagged messages. Analysis of Spanish revealed a split into two superdialects — an urban form shared across major cities, and a more regionally diverse form across rural areas and small towns.

- **Language Change**

  Tracking how varieties of a language evolve in space and time, combining the colloquial register of Twitter with the standard written register of Google Books to follow shifts in vocabulary and spelling.

- **The Global Language Network**

  Measuring the importance of a language from the structure of the network linking multilingual speakers and translated texts — book translations, Wikipedia editions, and Twitter — rather than from speaker counts alone.

- **Culture & Semantics**

  Inferring cultural regions directly from the words people use, without imposing ad-hoc traits, and quantifying how far shared meaning explains who talks to whom online.

## Mapping the Americanization of English

As global political preeminence shifted from the United Kingdom to the United States, so did the capacity to exert cultural influence. By analyzing a large corpus of geolocated tweets alongside the Google Books datasets for volumes published in the US and the UK, this line of work traces the spatial and temporal evolution of written English across both registers — the standard written language and the more colloquial forms of microblogging — and finds American English to be the dominant form outside the UK.

- [Read the paper: Mapping the Americanization of English in space and time](https://arxiv.org/abs/1707.00781)

## Key Publications

- **[American cultural regions mapped through the lexical analysis of social media](https://www.nature.com/articles/s41599-023-01611-3)**

  Louf, **Gonçalves**, Ramasco, Sánchez, Grieve. Humanities and Social Sciences Communications (2023)

- **[Mapping the Americanization of English in space and time](https://arxiv.org/abs/1707.00781)**

  **Gonçalves**, Loureiro-Porto, Ramasco, Sánchez. PLOS ONE (2018)

- **[Semantic homophily in online communication: evidence from Twitter](https://arxiv.org/abs/1606.08207)**

  Šćepanović, Mishkovski, **Gonçalves**, Nguyen, Hui. Online Social Networks and Media (2017)

- **[Learning Spanish dialects through Twitter](https://arxiv.org/abs/1511.04970)**

  **Gonçalves**, Sánchez. Revista Internacional de Lingüística Iberoamericana (2016)

- **[Topical differences between Chinese language Twitter and Sina Weibo](https://arxiv.org/abs/1512.07281)**

  Zhang, **Gonçalves**. NLPIT’16 — Workshop on Natural Language Processing for Informal Text (2016)

- **[Links that speak: the global language network and its association with global fame](https://bgoncalves.com/pdf/ronen-et-al-2014-links-that-speak-the-global-language-network-and-its-association-with-global-fame.pdf)**

  Ronen, **Gonçalves**, Hu, Vespignani, Pinker, Hidalgo. PNAS (2014)

- **[Crowdsourcing dialect characterization through Twitter](https://arxiv.org/abs/1407.7094)**

  **Gonçalves**, Sánchez. PLOS ONE (2014)

- **[The Twitter of Babel: Mapping World Languages through Microblogging Platforms](https://arxiv.org/abs/1212.5238)**

  Mocanu, Baronchelli, Perra, **Gonçalves**, Vespignani. PLOS ONE (2013)

A complete list is available on the [Publications page](https://bgoncalves.com/publications).

## Data & Code

Open-source tools, Python notebooks, and teaching materials covering natural language processing, word embeddings, and text analysis are published through Data For Science.

- [Explore the code at github.com/DataForScience](https://github.com/DataForScience)
