What question did this study set out to answer?

To construct a dynamic emotional evolution map tracking medical student burnout by integrating contextual sentiment recognition with latent topic discovery.

August 16, 2026Advanced ElectromagneticsOpen Access

Construction of Emotional Evolution Map of Online Texts on Medical Student Burnout Integrating BERT and LDA Topic Models

View Full Paper

Ask AI

Bookmark

Authors

JLJ. Y. Li

Discussion

Member takes

Overview

Machine learning study demonstrates dynamic tracking of burnout and stress topics in medical students from forum texts, highlighting scalable methods for psychological monitoring.

Key Points

To construct a dynamic emotional evolution map tracking medical student burnout by integrating contextual sentiment recognition with latent topic discovery.
Analyzed 137,430 online medical student text records using a domain-adapted BioBERT model to classify positive emotion, negative emotion, emotional exhaustion, and depersonalization.
Employed Latent Dirichlet Allocation (LDA) topic modeling to extract latent stress-related themes and quantify their distributions.
Combined sentiment trajectories with temporal topic intensity to characterize dynamic interactions across discussion themes.
The BioBERT classification framework achieved 92.7% accuracy with a macro-F1 score of 0.904.
The optimized LDA topic model attained a coherence score of 0.632.
Temporal analysis revealed distinct dynamic evolution patterns across internship stress, academic anxiety, career identity, and institutional support topics.

Cite This Study

J. Y. Li (2026) studied this question.

synapsesocial.com/papers/6a81799af2fb91fc834accf7 https://doi.org/10.7716/aem.v15i3.3319

View Full Paper

Ask AI

Bookmark

August 16, 2026Advanced ElectromagneticsOpen Access

Construction of Emotional Evolution Map of Online Texts on Medical Student Burnout Integrating BERT and LDA Topic Models

View Full Paper

Ask AI

Bookmark

Authors

JLJ. Y. Li

Discussion

Member takes

Overview

Machine learning study demonstrates dynamic tracking of burnout and stress topics in medical students from forum texts, highlighting scalable methods for psychological monitoring.

Key Points

To construct a dynamic emotional evolution map tracking medical student burnout by integrating contextual sentiment recognition with latent topic discovery.
Analyzed 137,430 online medical student text records using a domain-adapted BioBERT model to classify positive emotion, negative emotion, emotional exhaustion, and depersonalization.
Employed Latent Dirichlet Allocation (LDA) topic modeling to extract latent stress-related themes and quantify their distributions.
Combined sentiment trajectories with temporal topic intensity to characterize dynamic interactions across discussion themes.
The BioBERT classification framework achieved 92.7% accuracy with a macro-F1 score of 0.904.
The optimized LDA topic model attained a coherence score of 0.632.
Temporal analysis revealed distinct dynamic evolution patterns across internship stress, academic anxiety, career identity, and institutional support topics.

Cite This Study

J. Y. Li (2026) studied this question.

synapsesocial.com/papers/6a81799af2fb91fc834accf7 https://doi.org/10.7716/aem.v15i3.3319

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark