#wordembedding — Public Fediverse posts

This week for #TrainingTuesday, our friends over at Programming Historian have this lesson is designed to get you started with #WordEmbedding models 👀

➡️ Browse this resource: https://campus.dariah.eu/resource/posts/understanding-and-creating-word-embeddings

🎃 My #HALLOWEEN treat for you is a new #PUBLICATION 🎃

In #ComputationalCommunicationResearch, we present a semantic validation of using #WordEmbedding metrics to assess implicit group #stereotyping in text corpora: https://doi.org/10.5117/CCR2025.1.14.MULL

Last but not least, this validates our own study on "differential #racism in the #news", published in #PoliticalCommunication back in 2023: https://doi.org/10.1080/10584609.2023.2193146

@commodon @communicationscholars #openaccess #preregistered #opendata #openmaterials

New publication from our project on "Implicit and Explicit #Racism in #News and SocialMedia":

We explore the portrayal of #EthnicGroups in German legacy and #AlternativeMedia in 2022 - focusing on those group labels that were most frequently mentioned in the press last year.

The analysis is based on #WordEmbedding models that contrast the group labels' implicit and explicit associations with fear and admiration.

https://doi.org/10.25521/mzesfokus.2023.275

@commodon @communicationscholars #Journalism

#OutNow: My first co-authored article in a peer-reviewed journal. We investigate how different computational approaches fare with complex literary texts: https://doi.org/10.3389/fdata.2022.886362 (open access).

Summary in thread 🧵

@sociology @oneabstractaday #socialscience #methods #textanalysis #dictionary #wordembedding #scaling #machinelearning #sentiment #complex #literature