NLP Ecosystem in R #rstats #nlproc #textasdata #textmining #datamining #ML , Nice post here bnosac.be/index.php/blog…

Just finished the Text Data in #Economics course taught by Elliott Ash at the Warwick Economics Summer School 2022.

A fantastic public resource for those interesting in learning #textasdata , #NLP tools & #machinelearning methods.

You can freely access here: tinyurl.com/2e83z538

🎉 Beyond happy that this paper is finally out in Political Analysis (w/ twitterless Max Wich), which I started at the beginning of my PhD Political and Social Sciences at EUI #textasdata

doi.org/10.1017/pan.20…

An overview of the NLP ecosystem in R ( #nlproc #textasdata ) wp.me/pMm6L-K2K #rstats #DataScience

Published online today! Sol Hart Stuart Soroka and I measured politicization and polarization in early #COVID19 newspaper and network news coverage using #textasdata tools from Sven-Oliver Proksch Jon Slapin & Kenneth Benoit #quanteda journals.sagepub.com/doi/full/10.11…

My new article with Carsten Q. Schneider which scales the (il)liberalism of public discourses (speeches from 40 political leaders of 27 countries) using #rstats , #textasdata , #quanteda , and the Structural Topic Model of Molly Roberts et al., full-text view: rdcu.be/bBlQw

🚨Software Publication Alert🚨
This one is for you, text-analysis folks:

I am very happy to share the publication of my text annotation R-package 'handcodeR' 🥳, which is now available on CRAN. 🧵1/7

CRAN.R-project.org/package=handco…

#CRAN #rstats #TextAsData #TextAnnotation

Wanting to do #multilingual #TextAsData research but no one is willing to fund your Google Translate API requests? Don't despair, use UKP Lab's #easyNMT python package.

Free, high-quality machine translation in as little as three lines of code 🥳

Code 👉colab.research.google.com/drive/1m12_r8m…

In our new article @JEPP_Journal, Maximilian Haag, Constantin Kaplaner and I introduce the first version of the #EUPLEX dataset tandfonline.com/doi/full/10.10…, 50 free copies: tandfonline.com/eprint/7NPWDEZ… #polisciresearch #textasdata

Full house and nice #textasdata presentations at #ISPP2019 . W Dr. Julie Wronski Christian Pipal Martijn Schoonvelde

I ran the #MuellerReport through R/Quanteda Initiative to make a *BASIC* #TextAsData visualization. Here are the top 50 bigrams in the report (2-word phrases/ngrams of n=2). Some OCR/pre-processing fails, but check out the frequency of JR and the June meeting!

🚨 New preprint! 🚨 w/ William J. Brady & Manos Tsakiris:

We map the moral language of 39 U.S. presidential candidates to show how they are connected / differentiated by their use of moral rhetoric. #textasdata

osf.io/wn5rx

A [visual] thread on findings: 1/

Hey #EconTwitter !
Great seminar today at EIEF - Einaudi Institute for Economics and Finance by Michael McMahon on “Policymakers’ #uncertainty ”!

#econometrics #economics #monetarypolicy #fomc #textmining #textanalysis #inflation #textasdata

It is #tutorialthursday ! Learn how to use #rstats package #quanteda (by Kenneth Benoit) with the Manifesto Corpus. #textasdata manifesto-project.wzb.eu/down/tutorials…

A really inpiring presentation from Hauke Licht outlining a multilingual text embeddings method for classification of political documents.

(part of the amazing lineup at the #TextAsData workshop orgnized by Elliott Ash & Sascha O. Becker, ETH Center for Law & Economics, Monash SoDa Labs 🥤📊, CAGE @ Warwick)

I ran word embeddings models on the debates of the Can. HofC between 1945-2019 and looked for the most common neighbours for *elderly*. Language about old-age changed quite a lot since 1945. Any guess what words would be most common this year? 🙄 #covid #popaging #textasdata

tbh, I found this to be the hardest teaching year of my career: students were great, but it was tough. Well done if you got this far!

Materials for my 'Text as Data' #textasdata class (lectures+'flipped') below. Hope someone finds them helpful.
github.com/ArthurSpirling…

🚨Publication Alert🚨

How do right-wing populist user comments affect online deliberation?

Tjasa Turnsek and me show in Media and Communication that populist FB comments stir replies but impair their deliberative quality:
doi.org/jnx5

#polcomm #populism #css #textasdata

I am currently baking some exclusive #textasdata on tonight's French political debate between Macron and LePen 🤓