Developing a conceptual model for the blockchain domain using natural language processing methods

This paper presents approaches to conceptual model development using natural language processing (NLP) and topic modeling techniques. The research relies on several advanced algorithms such as BERTopic, KeyBERT and IBM’s unsupervised annotator, which are used to extract key terms and concepts from text data. These methods allow you to automatically identify hidden topics and relationships, greatly simplifying the analysis of large volumes of text and structuring information. Particular attention is paid to adapting algorithms to effectively build conceptual models in settings where pre-structured data is not available. Topic modeling methods demonstrate high accuracy and flexibility, making them suitable for developing educational programs and structured courses in various disciplines. The use of these methods can significantly improve the quality of text analysis, automate the process of creating educational materials, and provide a deeper understanding of complex thematic subjects. In addition, these techniques facilitate the development of adaptive models that can be used for a wide range of educational and analytical purposes. In the future, the use of such algorithms has the potential to transform the process of creating educational materials and training programs.