New generation of Web technology depends on the meanings of the Web content and unstructured data. Semantic Web aims computers to understand the data like human, instead of counting the frequencies of the words or detecting keywords. Therefore, the socio-technical studies, which make sense of Web content have an attention in informatics analytics.
This research presents the affective norms of Turkish words (Turkish Happiness Dictionary, THD) which is a set of psychological normative scores to measure an average level of happiness from Turkish online unstructured large-scale data. THD contains 345 words and their happiness scores in the Turkish language. THD provides researchers with standard materials through which they can automatically measure online happiness of Turkish large-scale data. THD can be used in in real-time big data analytics.
Users have different cybernetic emotions and behaviors on social platforms than real life. Thus the words they use differ. To measure the happiness of online large-scale data correctly, existing dictionaries (ANEW and others) are not enough. The average emotions of large-scale data will be different in a virtual environment. In addition, cultural differences in societies using different languages are not taken into consideration in dictionaries such as ANEW and adaptation. A dictionary like THD which is completely implemented on social networks should be used. THD is for Turkish language. THD specifically designed for the use in social networks. Words were obtained by web-text mining. THD includes emoticons (i.e. , , …) and the acronyms (i.e. LOL, ...).
