Teodora Vuković

Multimodal research at the University of Zurich

[Teodora Vuković @teodora_vuko]

New paper out
"A Corpus-Based Analysis of the Grammatical Status of Short Demonstratives in the Timok Dialect"
https://t.co/kkFhwYYMyt
open access https://t.co/aNfjML294g
@UZH_LiZZ

[Teodora Vuković @teodora_vuko]

My dissertation finally got published
"Empirical Approaches to Variation. The Case of Timok Variety of Torlak"
https://t.co/b4NEeGglb4

[Teodora Vuković @teodora_vuko]

Check out our latest preprint
"Comparative Analysis of Modality Fusion Approaches for Audio-Visual Person Identification and Verification" By Aref Farhadipour, Masoumeh Chapariniya, myself and Volker Dellwo
https://t.co/Zyw6annMnX
@cl_uzh @UZH_en

[Teodora Vuković @teodora_vuko]

Insula Care (by Tjade Tieltjes):

Call for Research Participants:

Study on Post-COVID Neurostimulation Therapy (Zürich)

https://t.co/hmqYSqxuNC

[Nagish Sign @NagishSign]

Making progress every day towards #SignLanguage translation! The future is inclusive and it's thrilling! #ArtificialIntelligence #Accessibility

[Teodora Vuković @teodora_vuko]

@iAmResearchApp Ah brilliant!!! I would have more of those if you need them!!! Noah you really don't seize to amaze me :)

[iAm: The Subjective Experience Research App @iAmResearchApp]

I just want to extend a major thank you to @meditative_dev for hosting the very first real-time reporting group session on iAm! So, what are we looking at here? We had seven people reporting on moments when they became aware that their attention had shifted away from their meditation object during a 5-minute session, but let's delve into more detail...

iAm enables self-reporting of any specific dimension of experience you wish to measure, in real-time as it unfolds. For our inaugural group session, we delved into the experience of distraction. Specifically, participants were asked to concentrate on a meditation object (such as the nose or breath) and press a button upon realizing their attention had deviated from the object.

This session served as a proof of concept, sparking many insightful discussions. Topics included defining the threshold between focused and distracted states, observations on how focus seems to improve in a group setting, and much more. The session was a learning experience, generating valuable lessons and suggestions for future endeavors.

The first image, directly from iAm, showcases a review page of the group session. Each time-series represents a unique participant's reported experience during the 5-minute session.

What's particularly exciting about iAm is how all this data is structured, stored within an underlying graph model. The second image displays the data as it appears in the database, with green nodes representing each user's unique session and blue nodes indicating moments when a participant became aware of distraction. The data structure is even more complex than what's visible, but showing all relationships would make the visualization overly busy.

Why is structuring this data as a graph beneficial? In short, it's all about RELATIONSHIPS! The more connections within the data, the greater the potential to uncover meaningful insights.

This is just the beginning, and the concept will continue to evolve as we identify other unique dimensions of experience worth exploring in this manner!

@danielmingram, @QualiaRI, @the_emergent, @ContemplateUoM, @RogerThisdell, any ideas come to mind :)?

[Christoph Hottiger @ChristophHotti1]

So this happened yesterday: I passed my viva and got to have a really stimulating discussion about my work with my wonderful supervisors @UZH_SPUR and @uzh_ds, Wolfgang Kesselheim and Heiko Hausendorf. Thanks to @HilokoKato for writing the protocol and to everyone who came to 🥳!

[Maximilian Krug @SelonFischer]

Computational and Quantitative Approaches to Multimodal Video Analysis - #CAMVA2023 here we go! I'm super excited 🤩

[Maximilian Krug @SelonFischer]

Geraldine Bengsch uses maschine deep learning to investigate "tst" in horse riding master classes. Super exciting! #CAMVA2023

[Maximilian Krug @SelonFischer]

André Krause presenting a very promising automatic anonymization approach relevant (not only) for #emca at #CAMVA2023 @KarolaPitsch

[IJCL @IJCL_journal]

In this #onlinefirst article, Teodora Vuković, Anastasia Escher, & Barbara Sonnenhauser of @UZH_LiZZ present a corpus-based method for assessing a range of dialect-standard variation, using a Torlak dialect corpus

#CorpusLinguistics
https://t.co/2P8LDP7EF5