Abstract:

In this study, it was aimed to conduct sentiment analysis of Turkish Twitter messages about autism between May 29, 2023 and June 2, 2023 through artificial intelligence applications. In the current study Turkish tweets related to autism were examined with the Python Tweepy application, which is one of the artificial intelligence applications, and the sentiment analysis of these tweets was performed. “Autism” is used as a keyword for scanning the tweets. By using the Bidirectional Encoder Representations from Transformers (BERT) method from transformers, feature extraction was performed, and the tweets were classified as positive and negative in two polarities with artificial neural networks. 1000 of the tweets about autism were randomly selected. Accordingly, it was determined that 66% of the tweets about autism contained positive and 34% of negative emotions. In this study, it was determined that Turkish tweets about autism, which were analyzed with artificial intelligence-based emotion analysis, often contained positive emotions. These messages are typically written by professional associations and mental health workers with the aim of raising awareness about autism, using simple sentence content targeting autism awareness.

Keywords: