The manuscript “Coping with Imbalanced Data in the Automated Detection of Reminiscence from Everyday Life Conversations of Older Adults” by Andrea Ferrario (Mobiliar Lab for Analytics at ETH and Chair of Technology Marketing, ETH) et al., has been accepted for publication in IEEE Access. The authors improve previous results on detection of reminiscence introducing a machine learning pipeline that comprises data augmentation with BERT (Bidirectional Encoder Representations from Transformer). Their results pave the way to interventions aiming at the improvement of the well-being of older adults.
