Identification of Ablaut and Rhyme Reduplications Using Fuzzy Inference System

Ablaut reduplication involves changing vowels, such as sing-song and chit-chat, while rhyme reduplications involve changing consonants such as hurly-burly. The occurrence of vowels in such reduplications follows certain patterns. To develop a fuzzy inference system that can predict the type of reduplications based on the phonetic patterns, this study examines the frequency of various phonetic patterns in the reduplications. This research used a quantitative design with statistical calculations and computer programming. The data were 150 samples reduplications, collected from textbooks, websites, and various internet sources. The samples were taken as isolated words, not in contexts (within sentences) as the meanings of the reduplications were not used as a variable in this research. The data were analyzed using chi-square for frequency analysis and a fuzzy inference system for the prediction. The results indicated a significant correlation between the frequency of occurrence and the type of reduplication. Moreover, rhyme reduplications have significantly more phonetic patterns than ablaut reduplications. Meanwhile, the system produces 92.9% accuracy in identifying the correct type of reduplication, and 80% accuracy in identifying whether an English word is a reduplication. This study implies a growing need for the application of computational linguistics in the future, that language can be studied from different perspectives, especially today, where there is a growing need for better natural language processing capabilities in computing.