newspapers get new authors and new guidelines in article writing. And the changes (apart from the en-dash) are too small to really spot with human eyes.
But I do agree with Mario that the 2025 change is... unexpectedly large.
Ok breakfast, then let's see about the quanteda package.
3/3 FOR NOW
But I do agree with Mario that the 2025 change is... unexpectedly large.
Ok breakfast, then let's see about the quanteda package.
3/3 FOR NOW
November 2, 2025 at 9:15 AM
newspapers get new authors and new guidelines in article writing. And the changes (apart from the en-dash) are too small to really spot with human eyes.
But I do agree with Mario that the 2025 change is... unexpectedly large.
Ok breakfast, then let's see about the quanteda package.
3/3 FOR NOW
But I do agree with Mario that the 2025 change is... unexpectedly large.
Ok breakfast, then let's see about the quanteda package.
3/3 FOR NOW
Text Analytics in R with quanteda (Part 1)
"Required Packages
library(quanteda)
library(quanteda.textstats)
library(quanteda.textplots)
library(readr)
library(dplyr)
library(ggplot2)
library(stringr)
library(DT)
library(tidytext)
Understanding Text Analytics Fundamentals
Text analytic...
Co..."
"Required Packages
library(quanteda)
library(quanteda.textstats)
library(quanteda.textplots)
library(readr)
library(dplyr)
library(ggplot2)
library(stringr)
library(DT)
library(tidytext)
Understanding Text Analytics Fundamentals
Text analytic...
Co..."
October 14, 2025 at 4:37 PM
Text Analytics in R with quanteda (Part 1)
"Required Packages
library(quanteda)
library(quanteda.textstats)
library(quanteda.textplots)
library(readr)
library(dplyr)
library(ggplot2)
library(stringr)
library(DT)
library(tidytext)
Understanding Text Analytics Fundamentals
Text analytic...
Co..."
"Required Packages
library(quanteda)
library(quanteda.textstats)
library(quanteda.textplots)
library(readr)
library(dplyr)
library(ggplot2)
library(stringr)
library(DT)
library(tidytext)
Understanding Text Analytics Fundamentals
Text analytic...
Co..."
Updates on CRAN: OmopSketch (0.2.0), pastboon (0.1.3), pim (2.0.4), qfa (4.0), quanteda (4.2.0), QuickJSR (1.5.1), RBesT (1.8-0), RcppGetconf (0.0.4), spatstat.explore (3.3-4), spData (2.3.4), statnet.common (4.11.0), tidyterra (0.6.2), webdav (0.1.3)
January 8, 2025 at 1:24 PM
Updates on CRAN: OmopSketch (0.2.0), pastboon (0.1.3), pim (2.0.4), qfa (4.0), quanteda (4.2.0), QuickJSR (1.5.1), RBesT (1.8-0), RcppGetconf (0.0.4), spatstat.explore (3.3-4), spData (2.3.4), statnet.common (4.11.0), tidyterra (0.6.2), webdav (0.1.3)
Updates on CRAN: bplsr (1.0.4), CDMConnector (2.1.1), data.table (1.17.8), future.mirai (0.10.1), gstat (2.1-4), MBAnalysis (2.1.0), newsmap (0.9.2), performance (0.15.0), quanteda (4.3.1), torch (0.15.1), ttservice (0.5.3), wordmap (0.9.5)
July 10, 2025 at 1:41 PM
Updates on CRAN: bplsr (1.0.4), CDMConnector (2.1.1), data.table (1.17.8), future.mirai (0.10.1), gstat (2.1-4), MBAnalysis (2.1.0), newsmap (0.9.2), performance (0.15.0), quanteda (4.3.1), torch (0.15.1), ttservice (0.5.3), wordmap (0.9.5)
I released the wordvector package v0.5.0. It is rapidly getting better and different from the original Word2vec package. Please read "Align word vectors of multiple Word2vec models" about the new function blog.koheiw.net?p=2299 #rstats #quanteda
Align word vectors of multiple Word2vec models
I have been developing a new R package called wordvector since last year. I started it as a fork of the Word2vec package but made several important changes to make it fully compatible with quanteda…
blog.koheiw.net
May 24, 2025 at 2:50 AM
I released the wordvector package v0.5.0. It is rapidly getting better and different from the original Word2vec package. Please read "Align word vectors of multiple Word2vec models" about the new function blog.koheiw.net?p=2299 #rstats #quanteda
Not sure if it's helpful but two R #packages which may be of use are quanteda:: (for quant text analysis) and tidytext:: is useful for text mining. If this helps at all? Otherwise I am clueless on the subject.
November 18, 2024 at 6:21 PM
Not sure if it's helpful but two R #packages which may be of use are quanteda:: (for quant text analysis) and tidytext:: is useful for text mining. If this helps at all? Otherwise I am clueless on the subject.
✨ New to NLP? We link a post that gives you a general understanding of text mining and the prominent bag of words approach in #rstats
https://bit.ly/text-mining-quanteda
https://bit.ly/text-mining-quanteda
Methods Bites
Blog of the MZES Social Science Data Lab
bit.ly
April 1, 2023 at 4:19 PM
✨ New to NLP? We link a post that gives you a general understanding of text mining and the prominent bag of words approach in #rstats
https://bit.ly/text-mining-quanteda
https://bit.ly/text-mining-quanteda
Quanteda, seu lindo. Adorei estudar e trabalhar com você hoje.
Amanhã será dia de só escrever análise e aprimorar os achados que tive no exploratório enquanto montava o código.
quanteda: Quantitative Analysis of Textual Data
Amanhã será dia de só escrever análise e aprimorar os achados que tive no exploratório enquanto montava o código.
quanteda: Quantitative Analysis of Textual Data
quanteda tutorials :: Tutorials for quanteda
Introduction to quantitative text analysis using quanteda
tutorials.quanteda.io
September 10, 2024 at 1:18 AM
Quanteda, seu lindo. Adorei estudar e trabalhar com você hoje.
Amanhã será dia de só escrever análise e aprimorar os achados que tive no exploratório enquanto montava o código.
quanteda: Quantitative Analysis of Textual Data
Amanhã será dia de só escrever análise e aprimorar os achados que tive no exploratório enquanto montava o código.
quanteda: Quantitative Analysis of Textual Data
If you've tried to use the #rstats NLP package {spacyr} in 2023, you may have noticed that the installation was broken. I fixed it and a new version including my changes is now on CRAN and GH 🥳
github.com/quanteda/spa...
github.com/quanteda/spa...
GitHub - quanteda/spacyr: R wrapper to spaCy NLP
R wrapper to spaCy NLP. Contribute to quanteda/spacyr development by creating an account on GitHub.
github.com
December 18, 2023 at 10:28 AM
If you've tried to use the #rstats NLP package {spacyr} in 2023, you may have noticed that the installation was broken. I fixed it and a new version including my changes is now on CRAN and GH 🥳
github.com/quanteda/spa...
github.com/quanteda/spa...
Se abordarán estrategias cuantitativas, cualitativas y computacionales para la investigación social, incluyendo:
🔹 IA y modelos de lenguaje
🔹 RRSS (Gephi, UCINET)
🔹 Análisis textual con R (Quanteda)
🔹 Etnografía multisituada
🔹 Participación social y datos digitales
🔹 IA y modelos de lenguaje
🔹 RRSS (Gephi, UCINET)
🔹 Análisis textual con R (Quanteda)
🔹 Etnografía multisituada
🔹 Participación social y datos digitales
June 5, 2025 at 8:19 PM
Se abordarán estrategias cuantitativas, cualitativas y computacionales para la investigación social, incluyendo:
🔹 IA y modelos de lenguaje
🔹 RRSS (Gephi, UCINET)
🔹 Análisis textual con R (Quanteda)
🔹 Etnografía multisituada
🔹 Participación social y datos digitales
🔹 IA y modelos de lenguaje
🔹 RRSS (Gephi, UCINET)
🔹 Análisis textual con R (Quanteda)
🔹 Etnografía multisituada
🔹 Participación social y datos digitales
If you think the number of topics, k, is the only important parameter for topic models, you need to read this post and the research paper. blog.koheiw.net?p=2233 I created a new model to optimize the Dirichlet priors to analyze imbalanced corpus more accurately. #rstats #quanteda
A new topic model for analysis imbalanced corpus
I have been developing and testing a new topic model called model Distributed Asymmetric Allocation (DAA) because latent Dirichlet allocation (LDA) takes a long time to fit to a large corpus but do…
blog.koheiw.net
November 23, 2024 at 10:02 AM
If you think the number of topics, k, is the only important parameter for topic models, you need to read this post and the research paper. blog.koheiw.net?p=2233 I created a new model to optimize the Dirichlet priors to analyze imbalanced corpus more accurately. #rstats #quanteda
I've used spacyr a lot. It is totally fine. Currently, there is a bug with the install process, but there is a well-working fix here :https://github.com/quanteda/spacyr/issues/236#issuecomment-1702311825
Reticulate has worked fine when needed. Should probably learn python though. Just, time...
Reticulate has worked fine when needed. Should probably learn python though. Just, time...
October 22, 2023 at 6:33 PM
I've used spacyr a lot. It is totally fine. Currently, there is a bug with the install process, but there is a well-working fix here :https://github.com/quanteda/spacyr/issues/236#issuecomment-1702311825
Reticulate has worked fine when needed. Should probably learn python though. Just, time...
Reticulate has worked fine when needed. Should probably learn python though. Just, time...
Updates on CRAN: crew (0.9.2), crew.cluster (0.3.1), IRR2FPR (0.1.1), keyATM (0.5.2), nimbleAPT (1.0.6), Qindex.data (0.1.1), quanteda (4.0.2), ShinyItemAnalysis (1.5.1)
April 24, 2024 at 5:15 PM
Updates on CRAN: crew (0.9.2), crew.cluster (0.3.1), IRR2FPR (0.1.1), keyATM (0.5.2), nimbleAPT (1.0.6), Qindex.data (0.1.1), quanteda (4.0.2), ShinyItemAnalysis (1.5.1)
Updates on CRAN: aides (1.3.3), CopulaREMADA (1.6.2), CPC (2.6.0), glmmrBase (0.8.1), LoopAnalyst (1.2-7), mas (0.3), MultivariateAnalysis (0.5.0), OenoKPM (2.4.1), prompter (1.2.0), quanteda (4.0.1), sharpshootR (2.3), spAbundance (0.1.3)
April 8, 2024 at 9:13 PM
Updates on CRAN: aides (1.3.3), CopulaREMADA (1.6.2), CPC (2.6.0), glmmrBase (0.8.1), LoopAnalyst (1.2-7), mas (0.3), MultivariateAnalysis (0.5.0), OenoKPM (2.4.1), prompter (1.2.0), quanteda (4.0.1), sharpshootR (2.3), spAbundance (0.1.3)
I already blogged today, so don't want to write more. But here is the code
![library(atrrr)
j_followers <- get_followers("jamoeberl.bsky.social", limit = Inf)
j_follows <- get_follows("jamoeberl.bsky.social", limit = Inf)
library(tidyverse)
library(ggwordcloud)
library(quanteda)
library(quanteda.textstats)
data <- bind_rows(
followers = j_followers,
follows = j_follows,
.id = "from"
)
bluesky_dfm <- data |>
filter(!is.na(actor_description)) |>
mutate(id = paste0(actor_handle, "_", from)) |>
distinct(id, .keep_all = TRUE) |>
corpus(
docid_field = "id",
text_field = "actor_description"
) |>
tokens() |>
dfm() |>
dfm_select(
"[[:alpha:]]",
valuetype = "regex"
) |>
dfm_remove(
stopwords()
)
keyness <- textstat_keyness(bluesky_dfm, which(docvars(bluesky_dfm, "from") == "follows")) |>
mutate(n_total = n_target + n_reference)
keyness |>
slice_max(order_by = n_total, n = 100) |>
ggplot(aes(
x = chi2, y = 0,
label = feature, size = n_total, colour = chi2)
) +
geom_text_wordcloud() +
scale_size_area(max_size = 7) +
scale_colour_gradient(low = "red", high = "blue") +
theme_minimal()](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ntd53albt5ffa4rgervvgibd/bafkreiaokcd34a2dbxux3q44e2lpgt54vmwe3bqv6jhldxi6tdva6joa64@jpeg)
November 24, 2024 at 3:04 PM
I already blogged today, so don't want to write more. But here is the code
You don’t. You just need a DFM, you can do that easily outside of quanteda by just creating a contingency table of count by token. I thought it would be easier. It probably wasn’t necessary.
August 22, 2025 at 7:12 PM
You don’t. You just need a DFM, you can do that easily outside of quanteda by just creating a contingency table of count by token. I thought it would be easier. It probably wasn’t necessary.
Topic Modeling: la modélisation thématique avec R, Quanteda... et ChatGPT #cran ourednik.info/maps/2024/05...
Topic Modeling: la modélisation thématique avec R, Quanteda… et ChatGPT
Ce tutoriel présuppose que vous avez fait vos premiers pas avec le module R Quanteda et que vous maîtrisez les notions stemming, stopwords, matrice document-terme, etc. On part du principe que vous av...
ourednik.info
November 20, 2024 at 8:58 PM
Topic Modeling: la modélisation thématique avec R, Quanteda... et ChatGPT #cran ourednik.info/maps/2024/05...
New conceptual review + tutorial on text embeddings out in #APA_Journals w/ @almogsi. Beginner-friendly, but experts will find spicy new takes as well. Tag a colleague who’s still counting words... #RStats #tidyverse #quanteda
1/
1/
June 24, 2025 at 2:10 PM
New conceptual review + tutorial on text embeddings out in #APA_Journals w/ @almogsi. Beginner-friendly, but experts will find spicy new takes as well. Tag a colleague who’s still counting words... #RStats #tidyverse #quanteda
1/
1/