Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Dokuz Eylül Üniversitesi, Fen Bilimleri Enstitüsü, Veri Bilimi, Türkiye
Tezin Onay Tarihi: 2024
Tezin Dili: Türkçe
Öğrenci: Buse Demir
Danışman: Özlem Ege Oruç
Özet:
In
this study, classification models were implemented using natural language
processing (NLP) methods
to predict individuals' mental distress. The study utilized
a labeled dataset of social media user comments, which included tags
indicating the users' mental health
status. Two different
word representation approaches, Word2Vec and Term Frequency-Inverse Document Frequency (TF-IDF),
were evaluated. To classify mental health status, models such as Naïve Bayes
(NB), Support Vector Machine (SVM), K-Nearest Neighbors
(KNN), and eXtreme Gradient Boosting (XGBoost) were applied. The results
indicated that the XGBoost model, developed with data represented by TF-IDF,
outperformed the other models.
The methods employed, the models applied, and the findings of
this study are expected to make significant contributions to further research
in the fields of machine learning, natural language processing, and mental
health. The results obtained may lead to advancements in predictive mental
health monitoring systems and facilitate early intervention and support for
individuals experiencing psychological distress.
Keywords: