Design of a Psychological Status Monitoring and Early Warning Platform for College Students Based on Deep Learning Models
Clicks: 72
ID: 312640
2026
Article Quality & Performance Metrics
Overall Quality
Not rated
Combines reader engagement with the AI quality analysis. This
article has not been analysed, so there is no overall score —
reader engagement is measured and shown alongside.
Reader Engagement
Steady Performance
21.3
/100
72 views
8 readers
AI Quality Assessment
Not analyzed
Readership in this journal
SteadyRanked #3 of 705 articles by views in Journal of Computing & Biomedical Informatics
Most read
Least read
Bar heights use a square-root scale. Only the 120 most-read articles are drawn; the journal has 705 in total.
Mint this article as an NFT
Not yet mintedCreate a permanent, verifiable on-chain record of this article on the Scimatic Network. The NFT is held in your Journament account, and you can withdraw it to your own wallet at any time.
5
SUSD
one-off · no wallet required
Abstract
The increasing prevalence of psychological problems among college students warrant the creation of smart and dependable early warning mechanisms. This paper presents a hybrid Psychological Early Warning Network (HPEW-Net) which is a new multimodal deep learning model that can forecast the level of psychological risk through heterogeneous data on the students. The suggested model will integrate organized academic and behavioral data, trends of activities, textual feedback, and social interaction data to give a complete picture of the student mental health. HPEW-Net uses modern deep learning systems, such as FT-Transformer to model tabular data, Temporal Fusion Transformer to analyze time-series, DeBERTa-v3 to understand contextual text, and Graph Neural Networks to understand social relationships. The modality-specific representations are fused using an attention based fusion mechanism that dynamically allocates relevance to the various sources of data thereby ensuring robust and adaptive learning. Experimental results demonstrate that the proposed model compares to the traditional machine learning and base deep learning methods in several measures, such as accuracy, precision, recall, and the F1-score. The model is found to be convergent and has a good performance in generalization, as seen through low discrepancies between training and validation performances. Further analysis of its capability to relate to various levels of psychological risk cases is supported by confusion matrix and heatmap which shows that it has the capability to correctly categorize various levels of critical misclassification of high-risk students. The results underscore the successfulness of mental health monitoring predictive learning by utilizing multimodal dimensions as well as attention-based fusion. The proposed framework is an efficient and scalable measure of early psychological risk identification and can be used in educational institutions to intervene in a timely manner and help students.
| Reference Key |
imported_1777055717_69ebb7e57764e
Use this key to autocite in the manuscript while using
SciMatic Manuscript Manager or Thesis Manager
|
|---|---|
| Authors | Wei Hao |
| Journal | Journal of Computing & Biomedical Informatics |
| Year | 2026 |
| DOI |
10.56979/1002/2026/943
|
| URL | |
| Keywords | Keywords not found |
Citations
No citations found. To add a citation, contact the admin at info@scimatic.org
Comments
No comments yet. Be the first to comment on this article.