Title: An evaluation model for college students' mental health based on machine learning algorithm

Authors: Qiaoying Ming

Addresses: School of Information Engineering, Xi'an FanYi University, Xi'an, Shaanxi, China

Abstract: Owing to the traditional beliefs, people tend to be hesitant and reserved in expressing themselves. To achieve accurate assessment of college students' mental health problems, a CNN-BiLSTM mental health assessment algorithm based on metaphorical attention mechanism is proposed. CNN-BiLSTM text processing module and metaphorical attention mechanism are used to improve the evaluation effect. The results show that compared with Text-CNN, BiLSTM+multi-layer RNN and BiLSTM+Attention, the recall rate and F1-value of the proposed algorithm are increased by 6.52% and 4.04%, respectively, and the prediction effect is best. After the elimination of RNN_MIP, metaphorical attention mechanism and BiLSTM, F1-value decreases by 2.33%, 8.72% and 5.7%, respectively, and the decrease is obvious.

Keywords: machine learning; mental health; social platform; evaluation and prediction; metaphorical feature.

DOI: 10.1504/IJWMC.2025.143017

International Journal of Wireless and Mobile Computing, 2025 Vol.28 No.1, pp.58 - 67

Received: 21 Nov 2023
Accepted: 15 Apr 2024
Published online: 02 Dec 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article