Developing a Comprehensive Emotion Lexicon for Turkish

An emotion lexicon is a tool that links emotion-related keywords in a language to comparable words in the language’s dictionary. While English and Spanish possess numerous lexicons dedicated to emotions, languages like Finnish and Turkish, characterized by complicated morphological structures do not exhibit the same level of lexical abundance. With this reason, this study aims to develop a comprehensive emotion lexicon tailored to the Turkish language. To start, a compilation of Turkish emotion words was generated specifically for this objective. After that, collection of sentences was assembled using 100 literary works written in the Turkish language. The construction of a matrix was carried out to demonstrate the co- occurrence of emotion terms and other words within sentences, employing the term frequency inverse document frequency (TF-IDF) method. Vectors corresponding to each emotion were generated individually. A cross-validation study was conducted to compare the emotion assignments from the study and ChatGPT. The results of the study indicated that the emotion lexicon displayed a high accuracy when applied to textual data. A consensus was reached by four individuals regarding the experience of at least one emotion among the five identified emotions, thus indicating successful recognition of the predominant emotion conveyed in the text.