The project focuses on the research of methods of analysis of the conversational content that is accumulated on
social media and contains different forms of toxicity. The project will propose approaches to the detection of toxicity of
various kinds by processing text data and visual data. The analysis of texts and images will be supplemented by speech processing from recordings
and videos,therefore the analysis will be multimodal. As part of the project, new methods of multimodal toxicity
detection will be proposed.The project will result in adaptive models that could significantly support the fight against misinformation in the online space and thus
contribute to the truth of the web.
1) research artificial intelligence methods with the aim of
finding detection models for multimodal data (text, image and speech)
2) monitor the development of themes and types of antisocial and toxic messages
3) recognize the types of authors of toxic content and the inconsistency of their
typology with presented content
4) investigate of ways to achieve that the studied models are explainable and
interpretable
5) ensure the applicability of the models also for the Slovak language and Slovak conditions.
All associated materials are available on the following links:
docent
doktorand
docent
docent
doktorand
doktorand
docent
profesor
profesor
doktorand
odborný asisent
odborný asisent
doktorand
doktorand
doktorand
profesor
docent
docent
doktorand
doktorand
odborný asistent
doktorand
výskumný pracovník - expert
výskumný pracovník - senior
výskumný pracovník - senior
výskumný pracovník - expert
výskumný pracovník - expert