text-as-data

Beautiful visualizations of how language differs among document types.

A python package to run contextualized topic modeling. CTMs combine contextualized embeddings (e.g., BERT) with topic models to get coherent topics. Published at EACL and ACL 2021 (Bianchi et al.).

Text analysis with networks.

Interpretable data visualizations for understanding how texts differ at the word level

Notebooks for the Seattle PyData 2017 talk on Scattertext

Closed captions from Internet Archive TV News, 2009–2023

Summer/ winter schools, workshops and conferences in computational social science 🫂

A tool for Semantic Scaling of Political Text (branch of Topfish, a suite of tools for Political Text Analysis)

CNN Transcripts 2000--2025

Literature 📄 and datasets 📚 on automatic populism detection

[PSRM 2025] LinkOrgs: An R package for linking linking records on organizations using half a billion open-collaborated records from LinkedIn

Code and models for 3 different tools to measure appeals to 8 discrete emotions in German political text

2018 Computational Text Analysis Notebooks, University of Mannheim

Summer 2017 Social Media Analytics Workshop Series

An Automation Webcrawler for Extracting Central Bankers' Speeches

'dictvectoR' measures the similarity between a concept dictionary and documents, using fastText word vectors. Implements the "Distributed-Dictionary-Representation" (Garten et al. 2018) method in R.

PartySOME: open datasets on political parties' social media, covering activity for 498 parties in 37 countries and monthly issue agendas for 219 parties in 18 European countries, to 2023.

The ABC of Computational Text Analysis. BA Seminar, Spring 2022, University of Lucerne

40+ validated topic models for computational social science, behind one NumPy-native interface: covariate effects on topic prevalence and content, shared diagnostics, reproducible fits. LDA, STM, CTM, keyATM, BERTopic, ETM, and more on a parallel Rust core. One wheel, no JVM, no PyTorch.

NBC-hosted MSNBC transcripts 2008–2014