Depression in Social Media: the Experience of the eRisk Lab of CLEF

Several studies have shown that the words people use are indicative, through distinctive linguistic patterns, of their psychological state. In 2017 this was the starting point of the eRisk initiative that aims at the early detection of signs of psychological problems from users' posts in social media. In this talk I will give an overview of the activities of eRisk in CLEF in the last four years with particular attention to the early detection of signs of depression. I will also discuss how we passed from the detection to the estimation of the severity of the symptoms of depression. I will conclude summarising the lessons learned and indicating the next steps of the initiative.

Fabio Crestani

Fabio Crestani is a Full Professor at the Faculty of Informatics of the Universitab  della Svizzera Italiana (USI) in Lugano, Switzerland, since 2007. Previously he was a professor at the University of Strathclyde in Glasgow, UK. He holds a degree in Statistics from the University of Padua (I) and a MSc and PhD in Computing Science from the University of Glasgow (UK). Prof Crestani worked in Information Retrieval, Text Mining and Digital Libraries for the past 30 years. In these areas he has published over two hundred refereed papers on both theoretical and experimental investigations. He has also served in the organising and program committees of several conferences and in the editorial boards of several journals.