Models of Memorability: Learning, Experiments, and Applications

We describe the construction and use of predictive models that provide inferences about the likelihood that users will consider particular events to be memorable landmarks in time. We discuss experiments and present integration of the models of event memorability into prototype file browsing tools. Finally, we discuss ongoing research and future directions for using predictive models of human memory in computing.