Research Information System
Thesis Type: Doctorate
Institution Of The Thesis: Gazi University, Fen Bilimleri Enstitüsü, Turkey
Approval Date: 2023
Thesis Language: Turkish
Student: Doğan KÜÇÜK
Supervisor: Nursal Arıcı
Open Archive Collection: AVESIS Open Access Collection
Abstract:
Today, social media is known to be an important source of data for many different fields. One of these fields is public health, especially the automatic monitoring of public health. On the other hand, there are rapid and profound developments in the field of artificial intelligence. The main methods within the scope of artificial intelligence are machine learning models and, particularly, deep learning models today. Additionally, subfields such as natural language processing and sentiment analysis, stance detection, and named entity recognition are also within the scope of artificial intelligence. In our thesis, traditional machine learning methods and deep learning methods were used for sentiment analysis, stance detection, and named entity recognition on social media posts for the automatic monitoring of public health. Comparative evaluation results were presented on annotated tweet datasets for Turkish. It has been observed that deep learning-based models achieve higher performance. Furthermore, within the scope of our thesis, a public health monitoring and decision support system utilizing the approaches solving the aforementioned problems was proposed. Our thesis is highly significant as it involves solving important natural language processing problems using machine learning and deep learning models on common datasets, creating datasets for Turkish natural language processing during this process, and making the proposed system usable by public health experts and decision-makers; hence it contributes significantly to the relevant literature in these aspects.
Key Words : Natural language processing, health informatics, deep learning, machine learning, sentiment analysis, stance detection, Turkish, social media