Naslov
Extractive summarization of scientific publications

Autori
Krušić, Lucija

Vrsta, podvrsta i kategorija rada
Ocjenski radovi, diplomski rad, diplomski

Fakultet
Odjel za informatiku

Mjesto
Rijeka

Datum
16.09

Godina
2019

Stranica
125

Mentor
Martinčić-Ipšić, Sanda

Ključne riječi
automatic text summarization, extractive summarization, abstractive summarization, natural language processing, deep neural networks

Sažetak
This Master’s thesis aims to provide a comprehensive survey of state-of-the art methods used for the task of Automatic Text Summarization. Automatically made summaries can provide great benefits to everyday internet users and enhance the way we search for relevant and necessary information and save time and resources invested in human-made summaries. This thesis provides an overview of the field of ATS and covers the approaches to summarization, the real-world applications of summarization and the various evaluation metrics used to establish the quality of the generated summary. Automatic text summarization is a blooming field which has recently gained significant interest and presently, much advancement is being achieved with the use of neural networks. The thesis will provide a comprehensive survey of the recent work done in the field, including the datasets used and the state-of-the art results of recent studies. The various methods of approaching ATS are described in depth and compared on the basis of their effectiveness.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo, Informacijske i komunikacijske znanosti

POVEZANOST RADA