Determination of Natural Language Processing Tasks and Tools for Topological Functioning Modelling

Topological Functioning Modelling (TFM) is based on analysis of exhaustive verbal descriptions of the domain functionality. Manual acquisition of knowledge about the domain from text in natural language requires a lot of resources. Natural Language Processing (NLP) tools provide automatic analysis of text in natural language and may fasten and make cheaper this process. First, the knowledge, its expressing elements of the English language, and processing tasks that are required for construction of the topological functioning model are identified. The overview of the support of these tasks by the main NLP pipelines is based on the available documentation without performing practical experiments. The results showed that among the selected six NLP pipelines the largest support comes from the Stanford CoreNLP toolkit, FreeLing, and NLTK toolkit. They allow analysing not only the words and sentences, but also dependencies in word groups and between sentences. The obtained results can be used for academics and practitioners that perform research on NLP for composition of domain (business, system, software) models.