PYTHON-BASED CREATION OF A DOMAIN DICTIONARY
DOI:
https://doi.org/10.34132/mspc2025.01.14.22Keywords:
domain vocabulary, RDF, NLTK, Python.Abstract
This abstract presents an analysis of the feasibility of creating a domain-specific dictionary using the Python programming language. Various data structures for storing the generated domain-specific dictionaries and Python tools for working with these structures are presented. A use case for a model with SPARQL query support is described. Software tools for text analysis and the extraction of identified terms are defined. A brief description of the proposed tools for term discovery is provided. The process of term discovery using natural language processing tools is described. An approach involving the search for pattern constructions (template matching) to determine the meanings of the identified terms is proposed. The applicability of regular expressions for searching and processing these pattern constructions is described. The option of using large language models for creating the domain-specific dictionary is considered.
References
Veselovskaya, G. V., & Lebed, O. S. (2018). Models of the subject field in computer multimedia projection technologies. Applied Questions of Mathematical Modeling, 2, 203–211. https://doi.org/10.32782/2618-0340-2018-2-203-211
Nayak, A., Božić, B., & Longo, L. (2022). Data Quality Assessment of Comma Separated Values Using Linked Data Approach. У Business Information Systems Workshops (p. 240–250). Springer International Publishing
Dongo, I., Cardinale, Y., & Chbeir, R. (2018). RDF-F: RDF Datatype inFerring Framework. Data Science and Engineering, 3(2), 115–135.
Nelli, F. (2018). Textual Data Analysis with NLTK. У Python Data Analytics (p. 487–506). Apress.
Cao, Z., & Ren, Q. (2024). Towards python program repair with generative pre-trained transformer (GPT-3.5). Theoretical and Natural Science, 43(1), 102–112.
