A large part of existing data is available in textual format, which mainly comes from human speech and coherent text. The processing of these kind of unstructured data sources can be provided by natural language processing. In case of morphologically rich languages it is inevitable to analyze words separately since the meaning is defined by the word form. The morphological reinflection is one of the most important tasks in natural language processing, which means generating an inflected form from a given source form and morphological paradigm. Generating words this way can be used for a wide range of use cases such as chatbots, search engines, data generation but it can also contribute to understanding extinct languages.
My thesis consists of two main parts. In the first one I define the conception and significance of the morphological reinflection and present the technologies I have used, as well as the solutions that have been made for this task so far. In the second half of the thesis, I describe in detail the language-independent morphological reinflection system I have made, the reasonableness of the planning decisions and the results obtained in Hungarian, Finnish, Turkish and Persian.