2022
DOI: 10.1016/j.chb.2021.107095
|View full text |Cite
|
Sign up to set email alerts
|

Machine learning for suicidal ideation identification: A systematic literature review

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
11
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6
2
1

Relationship

0
9

Authors

Journals

citations
Cited by 29 publications
(11 citation statements)
references
References 72 publications
0
11
0
Order By: Relevance
“…(1) A trend plot will visualize the progression of log file data publications [ 44 ]. (2) A taxonomy flow chart will showcase the complex relationships between various theories in different research domains [ 45 ]. (3) A summary table is crafted to encapsulate the characteristics of each study.…”
Section: Methodsmentioning
confidence: 99%
“…(1) A trend plot will visualize the progression of log file data publications [ 44 ]. (2) A taxonomy flow chart will showcase the complex relationships between various theories in different research domains [ 45 ]. (3) A summary table is crafted to encapsulate the characteristics of each study.…”
Section: Methodsmentioning
confidence: 99%
“…Comprehensive studies of learning-based techniques have been applied to digital data [6], medical record [7], complex and large (original) reports [8], social media text [9]- [11] and multimodal data [12]. Systematic studies and reviews towards this research field recapitulate the problem of demographic bias in data collection mechanism, managing consent of users for data, and theoretical underpinnings about human behavior in user-generated information [13]. Although such studies have made headway in human understanding of causes and perceptions associated with user-generated social media posts, automated identification of mental health related causes and perceptions is an area that has not yet come to fruition.…”
Section: A Current Position Of Communitymentioning
confidence: 99%
“…An increasing number of mobile solutions have been proposed in the literature for digital phenotyping [13]. Moreover, DL models have been intensively explored in combination with data from social media to detect suicidal ideation [67]. However, two research gaps remain: (1) there is a lack of solutions that deploy the models developed for suicidal ideation detection (different models have been developed but not deployed in solutions used in clinical settings) [23]; and (2) the passive monitoring of textual patterns is a too-little explored method in digital phenotyping studies [12,13].…”
Section: Contributions and Applicability Of The Boamentementioning
confidence: 99%