Розроблення інструментарію для аналізу текстів публічних та спеціалізованих джерел у завданнях передбачення та системного аналізу

• Видавець: The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
• Дата: 2020
• Автор: Savastiyanov, Volodymyr
• Формат: Стаття
• Мова: Ukrainian
• Опубліковано: The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2020
• Теми: системний аналіз; передбачення; text mining; NLP; класифікатори; онтології; OpenSource; Python; Gensim; системный анализ; предвидение; классификаторы; онтологии; systems analysis; foresight; classifiers; ontologies; Open Source
• Онлайн доступ: http://journal.iasa.kpi.ua/article/view/228316

A combined approach to extracting concepts and constructing classifiers and ontologies using open and proprietary software packages has been developed. Modern approaches, methods and models of storing large amounts of poorly structured information from Open Source software sets are studied. An ontology was built, in the leaves of which a classifier based on Boolean rules was implemented using SAS(R) Content Categorization Software. To build the ontology, the approach of constructing vectors of related concepts is employed using the Open Source library of Gensim software, namely the Word2Vec model. A typical algorithm for constructing a classifying ontology has been developed. The results of the research can be used to build an ontology of subject areas, create classification ontologies and mark corpora of texts.