Pia Pachinger
she / her
I'm a 3rd year PhD student in Natural Language Processing
working on making social media safer and more diverse. Particularly, I’m interested
in offensive / toxic text detection in online comment sections.
Who defines what is toxic? Often, what is considered as toxic is decided by online platforms themselves; and diverse perspectives on what is toxic to whom are overlooked. Developing toxicity detection systems with disaggregated labels is an interesting problem to tackle in my opinion. Therefore, I’m working on user-centric approaches to toxicity detection.
Natural language processing is still predominantly English-centred. I have frequently been working with Austrian German data that can be considered low-resource. In the future, I would like to extend my work to low-resource variants of Spanish.
I am lucky to have collaborated with political communication scientists throughout my PhD. I learned about rigorous methods to define toxicity and works measuring the effects of online toxicity which have been overlooked by computer scientists so far. I want to find ways to bring computer and social science closer.
I love all areas of Natural Language Processing, from foundations to application. Please reach out to me if you want to discuss!
Contact me by reordering @ tuwien.ac.at pia.pachinger
