UID:
almahu_9949372170802882
Format:
XII, 328 p. 70 illus., 50 illus. in color.
,
online resource.
Edition:
1st ed. 2022.
ISBN:
9783031044311
Series Statement:
Studies in Computational Intelligence, 1018
Content:
eRisk stands for Early Risk Prediction on the Internet. It is concerned with the exploration of techniques for the early detection of mental health disorders which manifest in the way people write and communicate on the internet, in particular in user generated content (e.g. Facebook, Twitter, or other social media). Early detection technologies can be employed in several different areas but particularly in those related to health and safety. For instance, early alerts could be sent when the writing of a teenager starts showing increasing signs of depression, or when a social media user starts showing suicidal inclinations, or again when a potential offender starts publishing antisocial threats on a blog, forum or social network. eRisk has been the pioneer of a new interdisciplinary area of research that is potentially applicable to a wide variety of situations, problems and personal profiles. This book presents the best results of the first five years of the eRisk project which started in 2017 and developed into one of the most successful track of CLEF, the Conference and Lab of the Evaluation Forum.
Note:
Early Risk Prediction of Mental Health Disorders -- The Challenge of Early Risk Prediction on the Internet -- A Survey of the First Five Years of eRisk: Findings and Conclusions -- From Bag-of-Words to Transformers: A Deep Dive into the Participation in the eRisk Early Risk Detection of Depression Tasks with Classical and new Approaches.
In:
Springer Nature eBook
Additional Edition:
Printed edition: ISBN 9783031044304
Additional Edition:
Printed edition: ISBN 9783031044328
Additional Edition:
Printed edition: ISBN 9783031044335
Language:
English
Subjects:
Computer Science
DOI:
10.1007/978-3-031-04431-1
URL:
https://doi.org/10.1007/978-3-031-04431-1
URL:
Volltext
(URL des Erstveröffentlichers)
Bookmarklink
