Research Article
A Machine Learning Approach to Mental Disorder Prediction: Handling the Missing Data Challenge
@INPROCEEDINGS{10.1007/978-3-031-63999-9_6, author={Tsholofelo Mokheleli and Tebogo Bokaba and Tinofirei Museba and Nompumelelo Ntshingila}, title={A Machine Learning Approach to Mental Disorder Prediction: Handling the Missing Data Challenge}, proceedings={Emerging Technologies for Developing Countries. 6th EAI International Conference, AFRICATEK 2023, Arusha, Tanzania, December 11--13, 2023, Proceedings}, proceedings_a={AFRICATEK}, year={2024}, month={6}, keywords={Data Imputation Machine Learning Mental Disorders Prediction Missing Values}, doi={10.1007/978-3-031-63999-9_6} }- Tsholofelo Mokheleli
Tebogo Bokaba
Tinofirei Museba
Nompumelelo Ntshingila
Year: 2024
A Machine Learning Approach to Mental Disorder Prediction: Handling the Missing Data Challenge
AFRICATEK
Springer
DOI: 10.1007/978-3-031-63999-9_6
Abstract
In recent years, the application of Machine Learning (ML) to predict mental disorders has gained significant attention due to its potential for early prediction. This study highlights the challenges of ML in mental disorders prediction, such as missing data in mental health datasets, by comparing four data imputation methods: Mode, Multivariate Imputation by Chained Equations, Hot Deck, and K-Nearest Neighbor (K-NN) to enhance predictive accuracy; and utilizing four ML classifiers and three ensemble methods: Bagging, Boosting, and Stacking, with Mode and K-NN imputation datasets to show consistent performance. The study ultimately contributes to early mental disorder diagnosis and intervention in alignment with the United Nations Sustainable Development Goal 3 (SDG 3) for global health and well-being, by highlighting ML and data imputation’s potential in mental health analysis and paving the way for further advancements in the field.
