Please note: this will be a hybrid-event hosted in the University of Bristol – registration is required via Eventbrite (click here to register), where you can select how you wish to attend. In-person places are limited so move quickly!
This masterclass will present an overview of recent work tackling self-imposed privacy issues as they arise in online conversations.
User engagement in online public discourse often includes self-disclosure — the revelation of personal information. Such disclosures on online public platforms (e.g., news forums) become a shared history, vulnerable to detrimental use by advertisers and malicious parties. Yet, users engage in self-disclosing behavior to attain strategic goals like relational development, social connectedness, identity clarification, and social control.
This masterclass will introduce the notion of self-disclosure as a construct to enable or facilitate online conversations and will present a taxonomy of the types of self-disclosure. We will then delve into the privacy implications of revealing personal information in unstructured conversations and on recent advances in automated detection of self-disclosure online.
We will discuss the role of self-disclosure in times of crises, highlighting as an example the incidence and evolution of self-disclosure temporally throughout the COVID-19 pandemic. Using a BERT-based supervised learning approach, we will share findings related to self-disclosure from a dataset of over 31 million COVID-19 related tweets. We will map users’ self-disclosure patterns, characterize personal revelations, and examine users’ disclosures within evolving reply networks. We uncover self-disclosure patterns in users’ interaction networks as they seek social connectedness and focused conversations during the pandemic.
We will conclude with open-ended research threads in this space. With the help of the audience we will discuss challenges and opportunities, including but not limited to the effects of influence and biases in self-disclosure, challenges with quantifying privacy risks, and the unequal distribution of privacy risks.
Please contact firstname.lastname@example.org with any questions – we look forward to seeing you there!