Ανάλυση συναισθήματος σε κειμενικά δεδομένα με χρήση ταξινομητών BERT
Sentiment analysis on textual data using BERT classification models
Bachelor Dissertation
Author
Κουρπάς - Δανάς, Αιμιλιανός
Date
2024-07Advisor
Σωτηρόπουλος, ΔιονύσιοςView/ Open
Keywords
Ανάλυση συναισθήματος ; Εξόρυξη γνώμης ; Βαθιά μηχανική μάθηση ; Μετασχηματιστές ; Επεξεργασία φυσικής γλώσσαςAbstract
Research in the field of sentiment analysis is an established field in natural language processing, seeking to identify the emotions expressed in a text. The effective application of sentiment analysis models can be a useful tool to better understand the needs, opinions, attitudes and preferences of the public.
To this end, the development of the models was based on the advanced transformer network architecture, with a focus on the specialized deep learning language model Greek-BERT. This approach enables the implementation, training and evaluation of machine learning models to accurately classify the content of a text as positive, negative or neutral. The central focus of the research is the application of the sentiment analysis technique to text excerpts in Greek, recorded from social media.