The application of the ChatGPT language model for automatic generation of structured abstracts
The application of the ChatGPT language model for automatic generation of structured abstracts

Summary/Abstract: Purpose/Thesis: The research aimed to evaluate the usefulness of the ChatGPT languagemodel in generating structured abstracts for academic publications.Approach/Methods: The methodology was qualitative. The study analysed ten articlesfrom the journal Zagadnienia Informacji Naukowej—Studia Informacyjne (5 in Polish,5 in English). ChatGPT version 4o was used to generate structured abstracts of the selectedarticles, then compared with the original abstracts to assess whether ChatGPT providedthe required information in each section.Results and conclusions: ChatGPT demonstrated strong capabilities in analysing andsummarising documents to create abstracts for scientific publications in the field of in-formation science. The language model performed well for both languages, with only twoabstracts exhibiting significant issues in specific sections.Originality/Value: The study showed the potential of language models, such as ChatGPT,in generating structured abstracts for bibliographic and full-text databases and as a com-plement to the researcher’s workshop.

• Details
• Contents