The paper reports on a research project aiming to facilitate automated abstracting in Hungarian language. In the framework of the project, the author developed an offline abstracting tool in Hungarian language. The efficiency of the first version of the programme was tested in comparison with summaries made by people. 340 people participated in the project, producing abstracts of articles from a wide range of scientific disciplines. The results of the project showed that the greatest differences were observed not between the automated summaries and the abstracts made by people, but between abstracts made by different people on the basis of the same article. The paper shows how the subject of the article affects the abstracting process, and draws useful conclusions regarding the process of automated abstracting. http://tmt.omikk.bme.hu/print.html?id=5209&issue_id=509
Financed by the National Centre for Research and Development under grant No. SP/I/1/77065/10 by the strategic scientific research and experimental development program:
SYNAT - “Interdisciplinary System for Interactive Scientific and Scientific-Technical Information”.