Overview of eRisk at CLEF 2023: Early Risk Prediction on the Internet (extended overview)
This article provides an overview of eRisk 2023, the seventh edition of the CLEF conference’s lab dedicated
to early risk detection. Our lab has been committed to exploring evaluation methodologies, effectiveness
metrics, and other associated processes in the field of early risk detection since its inception. The
applications of early alerting models are wide-ranging and span various domains, including health
and safety. eRisk 2023 encompassed three tasks. The initial task involved ranking sentences based on
their relevance to standardized depression symptoms. The second task concentrated on detecting signs
associated with pathological gambling early. Lastly, the third task required participants to automatically
estimate an eating disorders questionnaire by analyzing user writings on social media. In this extended
overview, we include additional details about the participants’ proposals and more detailed explanations
about metrics.
keywords: Early risk, Depression, Pathological gambling, Eating disorders
Publication: Congress
1697627003192
October 18, 2023
/research/publications/overview-of-erisk-at-clef-2023-early-risk-prediction-on-the-internet-extended-overview
This article provides an overview of eRisk 2023, the seventh edition of the CLEF conference’s lab dedicated
to early risk detection. Our lab has been committed to exploring evaluation methodologies, effectiveness
metrics, and other associated processes in the field of early risk detection since its inception. The
applications of early alerting models are wide-ranging and span various domains, including health
and safety. eRisk 2023 encompassed three tasks. The initial task involved ranking sentences based on
their relevance to standardized depression symptoms. The second task concentrated on detecting signs
associated with pathological gambling early. Lastly, the third task required participants to automatically
estimate an eating disorders questionnaire by analyzing user writings on social media. In this extended
overview, we include additional details about the participants’ proposals and more detailed explanations
about metrics. - Javier Parapar, Patricia Martín-Rodilla, David E. Losada, Fabio Crestani
publications_en