Making sure y'all saw this CfP for the Special Issue on Digital Communication in Social Movements.
Timeline: 8.3.25 abstract; 7.8.25 full paper; end of year publication.
#AI #TextasData #ImageAsData #SocialMovement #ScienceRocks
https://journal.computationalcommunication.org/announcement/view/179
Neu: Bei Pollux Political Corpora (PoliCorp) können umfangreiche politische Textsammlungen (Korpora) strukturiert durchsucht und analysiert werden.
Mehr Infos gibt es hier: https://pollux-fid.de/news/2025/79
Die Demoversion von PoliCorp kann hier getestet werden: https://lnkd.in/eCb3cgx5
Thinking about using #LLM for your #textasdata research? …Think twice! And join our panel tomorrow at #aoir2024
In March 2025, the RUNIP project @ruhr-uni-bochum.de will host the conference “Words in Numbers – Data-Driven Approaches to Texts in the Humanities and Social Sciences.” Keynote speakers include @jerielizabeth and Jo Guldi 🙌. The call for papers is now open, inviting especially early career researchers to submit proposals for short talks or posters. Check it out and join us! https://runip-projekt.ruhr-uni-bochum.de/words_in_numbers.html (English CfP is in the linked PDF at the end) #TextAsData #DigitalHumanities
“Text as Data”
From March 31 to April 5, 2025, a week-long
DH Spring School
in Potsdam, Germany will focus on text-based digital humanities for students and early-career scholars. Learn computational methods like topic modeling, stylometry, and network analysis. Apply by Nov 30, 2024. Participation is free!
The Language of (Non)Replicable Social Science #MetaScience #TextAsData
https://journals.sagepub.com/doi/full/10.1177/09567976241254037 There are systematic differences in the discussion of findings that were and were not replicable.
I was thinking that methods competence could be a confounder 1/
New post just out: running LLMs from my own laptop to rate how unhinged each of Medvedev's Telegram posts really is.
Actually, just testing the feasibility of using locally-deployed LLMs as coders of Russian-language text https://tadadit.xyz/posts/2024-04-09-inter-coder-reliability-unhinged-medvedev/ #Russia #TextAsData #LLM
Est-ce que vous connaissez des alternatives graphiques à #orgmode sur #emacs ?
J'ai beau être un #geek, j'ai bien les interfaces graphiques et la souris ^^
#textAsData
Papers about Causal Inference and Language
"A collection of papers and codebases about influence, causality, and language."
Stumbled over this pretty cool living bibliography on causal textanalysis. Might be useful to some. #textasdata #causal #nlp #sciencerocks https://github.com/causaltext/causal-text-papers
+++ NEW ARTICLE +++
Today I want to share some insights from my most recent publication – Text as Data: Variable Extraction via LSTM Networks!
#sociology #textAsData #CSS #AS @sociology
https://www.hendrik-erz.de/post/new-paper-text-as-data-variable-extraction-via-lstm-networks
Should we trust web-scraped data? https://arxiv.org/abs/2308.02231 When it comes to relationship btw target population and sampling, paper says "yes" if one takes care of volatility and personalization of web data plus incomplete indexing #TextAsData
Do you expect higher-quality #dialogue in small group discussions or whole class discussions?
It might depend on the metric:
- small groups fostered more invitation for peers to weigh in (d = 0.78, p < 0.001)
- whole classes generated more justifications of one's viewpoint (d = 0.69, p < 0.001)
Loads more insight from Herculean corpus analyses involving over 4000 students from 5 countries: https://doi.org/10.1016/j.linged.2023.101223
#edu #teaching #argumentation #P4C #corpusLinguistics #textAsData #DevPsych
Anyone have any particularly thorough walkthroughs re: looking for shared text across documents using R?
I am using the textreuse and text.alignment packages and making progress but I have been spinning my wheels trying to figure out exactly what I am looking at, as well as how the alignment score is calculated. I am interested in *both* what matches and what does not, often in large chunks (sentences at a time).
Alternately, anyone have thoughts on how to artfully present text alignment results, both for analysis and for sharing with others?
Thread:
🚨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=handcoge=handcodeR
Reminder: The #TextAsData conference deadline is now August 11th.
Details and link here: https://colliderbias.net/@commonsupport/110800066721843173
Gerade findet an der @elibbremen die 12. Sitzung des wissenschaftlichen Beirats von @fidpol statt. Wir sprechen über #neues #textasdata #openaccess #medienkompetenz #dataliteracy #computanionalsocialscience und mehr und freuen uns über den Austausch und die Expertise
Liberals and conservative use similar moral words, but they attach different meanings to those words
Maybe seems obvious, but suggests that moral politics is a competition over *meaning* and not promoting specific values
It also has implications for how we think about and use (or don't use) dictionary methods for understanding morality in text and speech.
#polisci #newresearch #newpaper #morality #socpsych #newpsychresearch @socialpsych @politicalscience #textasdata
If you are interested in how to apply #TopicModeling to continuously growing corpora, have a look at this talk by @rieger he gave last year in our lecture series:
https://zpid.cloud.panopto.eu/Panopto/Pages/Viewer.aspx?id=03addb74-2f86-4ed0-b877-af9c00a2959d
He shows the advantages of modeling growing text corpora using RollingLDA and presents some example applications, e.g., the uncertainty perception and inflation perception indicators, PsychTopics as well as various change detection and monitoring scenarios.
The Summer Institute in #ComputationalSocialScience will be held in person at the WZB #Berlin #SocialScience Center, July 3 to July 13, 2023
https://sicss.io/2023/berlin/
The focus of the SICSS-Berlin is on #TextAsData, #WebsiteSscraping, #MachineLearning, and #ethics.
Application deadline: March 31, 2023
If I were a #PhDStudent or #PostDoc I would apply
