Suicidal Ideation: Inferring Suicidal ideation: A new approach

Thumbnail Image

Files

URL

Journal Title

Journal ISSN

Volume Title

Perustieteiden korkeakoulu | Bachelor's thesis
Electronic archive copy is available locally at the Harald Herlin Learning Centre. The staff of Aalto University has access to the electronic bachelor's theses by logging into Aaltodoc with their personal Aalto user ID. Read more about the availability of the bachelor's theses.
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2024-09-20

Department

Major/Subject

Data Science

Mcode

SCI3095

Degree programme

Aalto Bachelor’s Programme in Science and Technology

Language

en

Pages

26 + 8

Series

Abstract

Suicide is a leading cause of death worldwide. Past research has focused on quantifying suicidal ideation through the use of questionnaires and interviews. In recent years, techniques such as machine learning have emerged as promising tools for predicting suicidal behaviour. This study aims to study the currently used suicidal ideation inference tools, and to investigate the usage of machine learning as a potential tool for inferring suicidal ideation using Patient health questionnaire-9 (PHQ-9) responses. PHQ-9 is a depression screening tool, used both in clinical and research settings. The study compares various machine learning models, namely Logistic regression, Random forest, Support vector machine, Gradient boosting, and Neural network. These models were trained and tested on demographical and web data with PHQ-9 scores as labels. Additionally, techniques such as feature engineering and Synthetic Minority Over-sampling (SMOTE) were used to manage class imbalance and identify key predictors for suicidal behaviour. The results indicate that no existing suicidal ideation inference tool can fully capture suicidal behavior in all scenarios. The machine learning models achieved accuracies of 0.61 (Logistic regression), 0.79 (Random forest, Gradient Boosting), 0.59 (Support vector machine), 0.75 (Neural network) and 0.70 (SMOTE random forest). The results also suggest that while machine learning offers valuable insights, it is unable to replace conventional methods for detecting suicidal ideation. However, the investigation provides a novel framework for future research by using passive web usage traces to infer suicidal ideation. This approach may enhance early identification and intervention efforts in mental health care.

Description

Supervisor

Korpi-Lagg, Maarit

Thesis advisor

Aledavood, Talayeh

Keywords

suicidal ideation, suicide, PHQ-9, machine learning, web data

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202411197185

Collections

[kand] Perustieteiden korkeakoulu / SCI