Abstract: Misinformation is rampant in online social networks. It is particularly prominent in information spaces where facts evolve rapidly, such as COVID-19. In this talk I will discuss common definitions of misinformation, how they are operationalized, and effective methods for their detection and characterization. Much of this talk will focus on the spreaders and receivers of misinformation. Spreaders are those accounts that work tirelessly to promote misinformation online. I will discuss how the built-in redundancies of their efforts enable their automated discovery. Next, we will discuss how individual differences of the misinformation targets can increase their susceptibility. Particularly, the level of polarization of these targets plays an important role in their susceptibility. We will discuss our recent work on understanding the process by which a particular person becomes polarized. We will conclude with a discussion of key challenges in the area of misinformation, and opportunities for future research.
Fred Morstatter is a Research Assistant Professor at the University of Southern California and a Research Lead at USC’s Information Sciences Institute.
His recent research has focused on forecasting, scientific reproducibility, and uncovering cultural models. Recently, he developed a geopolitical forecasting system that combined human and machine intelligence. This system won a year-long challenge under IARPA’s Hybrid Forecasting Competition. He has published over 40 papers on online dynamics including sampling bias in online systems, misinformation and online abuse. He served as the PC Co-Chair of ICWSM 2019. For more information, visit his web site at https://isi.edu/~fredmors.