Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTE
The growing demand for telemedicine has highlighted the need for automated healthcare services, particularly in medical question classification. This study presents a deep learning model designed to address key challenges in telemedicine, including class imbalance and accurate routing of Arabic medi...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-04-01
|
| Series: | AI |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2673-2688/6/4/77 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849712235823235072 |
|---|---|
| author | Bushra Al-Smadi Bassam Hammo Hossam Faris Pedro A. Castillo |
| author_facet | Bushra Al-Smadi Bassam Hammo Hossam Faris Pedro A. Castillo |
| author_sort | Bushra Al-Smadi |
| collection | DOAJ |
| description | The growing demand for telemedicine has highlighted the need for automated healthcare services, particularly in medical question classification. This study presents a deep learning model designed to address key challenges in telemedicine, including class imbalance and accurate routing of Arabic medical questions to the correct specialties. The model combines AraBERTv0.2-Twitter, fine-tuned for informal Arabic, with Bidirectional Long Short-Term Memory (BiLSTM) networks to capture deep semantic relationships in medical text. We used a labeled dataset of 5000 Arabic consultation records from Altibbi, covering five key medical specialties selected for their clinical relevance and frequency. The data underwent preprocessing to remove noise and normalize text. We employed stratified sampling to ensure representative distribution across the selected medical specialties. We evaluate multiple models using macro precision, macro recall, macro F1-score, weighted F1-score, and G-Mean. Our results demonstrate that DeepSMOTE combined with cross-entropy loss achieves the best performance. The findings offer statistically significant improvements and have practical implications for improving screening and patient routing in telemedicine platforms. |
| format | Article |
| id | doaj-art-60c73cc044a14b359ff94de776a42d08 |
| institution | DOAJ |
| issn | 2673-2688 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | AI |
| spelling | doaj-art-60c73cc044a14b359ff94de776a42d082025-08-20T03:14:20ZengMDPI AGAI2673-26882025-04-01647710.3390/ai6040077Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTEBushra Al-Smadi0Bassam Hammo1Hossam Faris2Pedro A. Castillo3King Abdullah II School of Information Technology, The University of Jordan, Amman 11942, JordanKing Abdullah II School of Information Technology, The University of Jordan, Amman 11942, JordanKing Abdullah II School of Information Technology, The University of Jordan, Amman 11942, JordanDepartment of Computer Engineering, Automatics and Robotics, Higher Technical School of Computer Sciences and Telecommunications Engineering (ETSIIT)-Communication and Information Technologies Researching Centre (CITIC), University of Granada, 18071 Granada, SpainThe growing demand for telemedicine has highlighted the need for automated healthcare services, particularly in medical question classification. This study presents a deep learning model designed to address key challenges in telemedicine, including class imbalance and accurate routing of Arabic medical questions to the correct specialties. The model combines AraBERTv0.2-Twitter, fine-tuned for informal Arabic, with Bidirectional Long Short-Term Memory (BiLSTM) networks to capture deep semantic relationships in medical text. We used a labeled dataset of 5000 Arabic consultation records from Altibbi, covering five key medical specialties selected for their clinical relevance and frequency. The data underwent preprocessing to remove noise and normalize text. We employed stratified sampling to ensure representative distribution across the selected medical specialties. We evaluate multiple models using macro precision, macro recall, macro F1-score, weighted F1-score, and G-Mean. Our results demonstrate that DeepSMOTE combined with cross-entropy loss achieves the best performance. The findings offer statistically significant improvements and have practical implications for improving screening and patient routing in telemedicine platforms.https://www.mdpi.com/2673-2688/6/4/77DeepSMOTEmulti-class classificationoversampling techniquesmedical questionsArabic language |
| spellingShingle | Bushra Al-Smadi Bassam Hammo Hossam Faris Pedro A. Castillo Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTE AI DeepSMOTE multi-class classification oversampling techniques medical questions Arabic language |
| title | Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTE |
| title_full | Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTE |
| title_fullStr | Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTE |
| title_full_unstemmed | Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTE |
| title_short | Enhancing the Classification of Imbalanced Arabic Medical Questions Using DeepSMOTE |
| title_sort | enhancing the classification of imbalanced arabic medical questions using deepsmote |
| topic | DeepSMOTE multi-class classification oversampling techniques medical questions Arabic language |
| url | https://www.mdpi.com/2673-2688/6/4/77 |
| work_keys_str_mv | AT bushraalsmadi enhancingtheclassificationofimbalancedarabicmedicalquestionsusingdeepsmote AT bassamhammo enhancingtheclassificationofimbalancedarabicmedicalquestionsusingdeepsmote AT hossamfaris enhancingtheclassificationofimbalancedarabicmedicalquestionsusingdeepsmote AT pedroacastillo enhancingtheclassificationofimbalancedarabicmedicalquestionsusingdeepsmote |