Machine learning in exploration the decoupling paradigm in transport

Abstract

The article deals with one of the paradigms of sustainable development, i.e. decoupling. The article focuses on machine learning in the field of text mining on decoupling in transport. The research problem was formulated in the form of a question: does machine learning improve the process of exploring information contained in scientific publications on decoupling in transport? The article aims to explore the decoupling paradigm in transport in tasks of natural language processing with the use of machine learning. In the article, research was conducted in the field of text mining and emotion profiler (by Plutchnik emotions), using lemmatization and techniques such as word cloud, a bag of words, Latent Dirichlet Allocation, RAKE (rapid automatic keyword extraction) algorithm, and hierarchical clustering. Important keywords, phrases for the description of the decoupling paradigm, methods of its research, and words that induce increased trust in this concept were found. The authors proposed an original framework for the research procedure.