Some methods for detecting & visualizing semantic change.
| Home | About | Resources |
Some perspectives on Xochitl Torres Small's win in CD NM-02
A congressional election success story, and an introduction to <nmelectiondatr>, an R data package that makes available New Mexico general election returns from 2014 to 2018.
Twitter, political ideology & the 115th US Senate
Estimating political ideologies of US Senators via the news media they share on Twitter.
Political ideology & New Mexico's 53rd State Legislature
A demonstration of the NOMINATE/VoteView procedure using roll call data from NMSL53.
New Mexico's 53rd State Legislature
A demo of the data package <nmlegisdatr>, which makes available data about New Mexico state legislative activities, including bill summaries, roll calls, & legislator details.
psychological and geographical distance in text
An investigation into the relationship between psychological distance and geographical distance in text.
building historical socio-demographic profiles
A simple workflow for building census-based, historical socio-demographic profiles
place from text: geography & distributional semantics
Some different methodologies for exploring the geographical information found in text.
topic models for synchronic & diachronic corpus exploration
A simple workflow from annotated corpus to topic model, with a focus on the exploratory utility of topic models.
a simple framework for corpus-based keyphrase extraction
A simple framework for identifying and extracting keyphrases from component texts of a corpus.
corpus query and grammatical constructions
A demonstration of the use of a simple collection of functions from my R-package `corpuslingr`.
a census-based approach to spanish language maintenance
An investigation into Spanish language maintenance within Hispanic communities in the US utilizing data from the US Census.
Building a historical, genre-based corpus Building a Naive Bayes classifier Model assessment & confusion matrix Summary In this short post, we outline a Naive Bayes (NB) approach to genre-based text classification. First, we introduce & describe a corpus derived from Google News’ RSS feed, which includes source and genre information. We then train, test & evaluate the efficacy of an NB classifier applied to online news genres, with some fairly nice results.