Selina Meyer is a PhD student and research assistant at the Chair of Information Science, University of Regensburg. Her research focuses on empathetic conversational agents for behaviour change. Currently, she receives a scholarship from the German Academic Scholarship Foundation.
The seventh ACM SIGIR Conference on Human Information Interaction and Retrieval (CHIIR) was supposed to happen in my hometown Regensburg, Germany.Unfortunately, it had to be moved to the virtual space due to the pandemic.To recreate the feel of the historic city, the local organisers took it upon themselves to model the city in gather town, complete with livestreams of popular places in town and Bavarian music. This was my second year attending CHIIR and both times it had to take place virtually, but the open and welcoming nature of the community made it easy to get to know others even in the virtual space and even for a relative newcomer like me.
The conference took place between Monday, 14th and Friday, 18th of March, starting with workshops and tutorials on Monday. We had a choice of four workshops and three tutorials of which I took part in three. The “First Early Career Researcher’s Roundtable for Information Access Research” organised by Johanne Trippas and David Maxwell invited early-stage researchers to reflect on the opportunities and challenges of entering academia during the pandemic. Over the course of the workshop, we identified lessons learned from the past years and defined recommendations for researchers and institutions to accommodate early career researchers (even) better in the future. The workshop’s results will be submitted to SIGIR Forum. “Catching Information Behaviour in the Wild” reflected on different ways to research behaviour in naturalistic and largely unregulated settings. It ended with brainstorming sessions in small groups which invited us to design our own approaches to catch information behaviour in the wild based on real case studies. “Conversational Information Seeking: Theory and Evaluation” gave an overview of the system design and evaluation of conversational information seeking from different perspectives. Other events focused on interdisciplinary, information quality and specific research topics within IR and Information Interaction.
The main program began on Tuesday. The program was staggered with repeated sessions to make each session accessible for people all over the world, so for me the day started at 9 am with a session on Conversational Search and Information Seeking, directly followed by two papers on Information Behaviour and Search-as-Learning. In total, we got to see 22 full papers distributed over 7 paper sessions and 14 short papers. Just as last year, what really stood out were two delightfully creative paper presentations by Dana McKay, George Buchannan, and their co-authors on the reasons behind filter bubble bursting and reactions to misinformation online. They exploited the virtual format in a way that highlighted its various opportunities and potential advantages compared to traditional presentation formats. Other presentations that stuck to my mind were Paul Thomas et al.’s reflections on how to better accommodate human factors in crowdsourcing, Chiraq Shah and Emily M. Bender’s warnings of giving too much power to large pretrained language models in search systems and Markus Bink et al.’s research which found that featured snippets tend to change users’ opinions on a topic, regardless of their correctness. The best paper award went to Tim Draws and co-authors, who proposed a two-dimensional viewpoint model as a better means to analyse user interaction in the context of debated topics.
Of course, the social events were the star of the show: In a pub-quiz style setting, participants competed in teams and performed information retrieval tasks about the history of Regensburg to score the most points. A virtual Bavarian breakfast with a live streamed tutorial of white sausage cooking and wheat beer pouring followed the day after. Everyone participating from their own home or office led to introductions to various traditional breakfast foods from people’s respective countries. Although the conference was fully virtual, these events helped create a sense of community and an opportunity for networking.
After the host of CHIIR 2024 – Sheffield – was revealed in the CHIIR it forward session and we got a sneak preview of the city – It looks stunning – the end of the conference for me was marked by the Doctoral Colloquium. There, I presented my dissertation project aimed at facilitating health behaviour changes and was introduced to the projects of four other talented PhD students. While the presented topics were varied and ranged from user-centric to system focused, all of them shared the property of focusing on social good: introduced projects were about improving information seeking for people affected by aphasia, creating a look-up system for sign language, exploiting augmented reality to increase the efficiency of literary research in neuroscience, and facilitating self-tracking for dementia patients. I am grateful to all the experienced researchers and mentors present at the DC for their insights and constructive criticism, that I’m sure will drive all our projects forward.
Overall, CHIIR 2022 was fun, thought provoking and inspiring. I thoroughly enjoyed meeting so many creative, and highly motivated people, the majority of which use their research abilities for social good by aiming to facilitate access to high quality information for all and minimise the spread of misinformation in various ways, a research goal that has become all the more important in recent years