Jelena Prokic relies on computational methods to explain how and why language changes and discover language internal and external factors that lead to observed change. She specializes in variation on the micro-level, where she is particularly interested in automatic dialect detection, automatic feature extraction and dynamics of language variation. She investigates problems related to formation, maintenance and evolution of dialects and their stability through time. She is dedicated to developing digital databases, tools and methods that would assist researchers in contributing to a theory of natural language and shed more light on language evolution, and eventually on human cognition, society and history. In her work she applies methods from natural language processing, machine learning, corpus linguistics and GIS.
Jelena holds a PhD in Computational Linguistics from the University of Groningen, MA degree in Computational Linguistics from the University of Tübingen and Diploma in Chinese Language and Literature from the University of Belgrade. After her PhD, she worked as a postdoctoral researcher at the University of Munich and the University of Marburg. She participated in various projects where she explored the potentials of digital methods and tools in the humanities. Her work includes projects on quantitative approaches to diversity of Bulgarian dialects (Buldialect), phylogeny of Native South American languages (QuantHistLing), digitization of German dialect data (Alignments of the PAD), automatic detection and analysis of the spread of neologisms in English (Incipient diffusion of lexical innovations) and development of the first digital phonetic database for computational historical linguistics and dialectometry (BDPA).