Open AccessEmotional Harmony in My Social Network
Author(s) -
Kaan Ant,
Banu Di̇ri̇
Publication year - 2018
Publication title -
akıllı sistemler ve uygulamaları dergisi
Language(s) - English
Resource type - Journals
ISSN - 2667-6893
DOI - 10.54856/jiswa.201812038
Subject(s) - happiness , closeness , sadness , anger , feeling , social media , turkish , sentiment analysis , psychology , microblogging , sentence , harmony (color) , computer science , social psychology , artificial intelligence , linguistics , world wide web , mathematics , mathematical analysis , philosophy , art , visual arts
Microblog has become a very popular communication tool among internet users these days. Microblog sites are a great data mine for natural language processing. Studies on these platforms are insufficient to determine the emotion in texts based on opinion analysis (positive, negative, neutral). In this study, it was shown how natural language processing can be used in sentence recognition in texts and a tool for recognizing emotion in Turkish language has been developed. The texts written by Twitter users are quoted and tagged with tweets that contain feelings of "happiness, fear, anger, sadness". Labeling was done using "left -k" and "binary vector" methods. In both methods, the "happiness" and "fear" classes are labeled with high achievements, but the achievements of the "anger" and "sad" classes are lower because of the closeness of the two senses.