A myriad of studies have proposed verification techniques for rumors spreading in social media that mainly relied on evidence from the propagation networks, user interactions, or external evidence from the Web. However, none of these studies exploited evidence from trusted authorities. This talk addresses the problem of authority finding along with detecting the stance of these authorities towards rumors spreading in Twitter.

For the first problem, we released the first test collection for Authority FINding in Twitter (AuFIN) and proposed a hybrid model that employs pre-trained language models and combines lexical, semantic, and network signals to find authorities. 

For the second problem, we constructed and released the first Authority STance towards Rumors (AuSTR) dataset. Due to the relatively limited size of our dataset, we studied the usefulness of existing stance detection datasets for our task and showed that existing datasets are somewhat useful for the task; however, they are clearly insufficient, which motivates the need to augment them with annotated data constituting stance of authorities from Twitter. 

Biography:

Fatima Haouari is a Computer Science PhD candidate at Qatar University (QU). She received her BSc and MSc. degrees in Computer Science from QU. Her research interests span information retrieval, natural language processing, and social media analytics. Fatima joined bigIR research group at QU in 2019, started her PhD since then, and was awarded a GSRA research grant from QNRF to support her PhD study. Her PhD focuses on evidence-based early rumor verification in Arabic Twitter. She published her work in top conferences such as ECIR and SIGIR, and journals such as IP&M. During her PhD journey, Fatima led the construction and annotation of multiple Twitter datasets in this area, which were released to the research community. She has co-organized CLEF CheckThat! Lab 2020, 2021, and 2023 targeting multiple problems related to rumor verification and fake news. She received the Best Paper Award in WANLP 2021 workshop for her work on Arabic COVID-19 rumor verification in Twitter.