Tools
Tools
-, Mohammad Farooque Azam (2023) Multi scale context aware drug review sentiment analysis using pretrained med-BERT.de. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (752kB) | Preview |
Preview |
PDF (Configuration manual)
Download (3MB) | Preview |
Abstract
A massive amount of drug-related data is available online in the form of reviews and comments. This data could be leveraged in analyzing the opinion and sentiments of users related to the drug. The written medical texts of patients and carers, in particular, have a considerable influence on how people, physicians, and drug developers regard drug users. Sentiment analysis approaches have grown from fundamental concepts to advanced methods of machine learning such as deep learning, which has emerged as a recent development in many NLP applications. The emergence of transformer-based models has revolutionized the field even further.. However, most of them have worked on the dataset from a single source with a limited number of rows and a general-purpose model for analysis, making them incapable of capturing the contextual meaning of medical terms used in the text. This defect affects the performance of the pharmaceutical recommendation system. Hence, this study proposes sentiment analysis of drug reviews using the medical domain-specific pre-trained transformer-based model medBERT.de. The proposed model achieved the highest F-score of 83%, a precision of 84% for optimal hyperparameter values of epochs=3, learning rate= 2e-5, batch size=32 and decay_weight= 0.1 , suggesting the model works fine in capturing the medical terms used in the dataset.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Trinh, Anh Duong (Senja) UNSPECIFIED |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science R Medicine > RS Pharmacy and materia medica P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Tamara Malone |
| Date Deposited: | 07 Nov 2024 15:45 |
| Last Modified: | 07 Nov 2024 15:45 |
| URI: | https://norma.ncirl.ie/id/eprint/7162 |
Actions (login required)
 |
View Item |