Multidocument Summarization

Student: Burshtein Denis

Supervisor: Ekaterina Artemova

Educational Programme: Applied Mathematics and Information Science (Bachelor)

Year of Graduation: 2020

Over the past few years, there has been an increased interest in applying neural networks to various Natural language processing tasks, such as: machine translation, document summarization, part-of-speech tagging etc. In recent years, neural networks allow achieving more and more significant results in this area. In this work, their application to the multi-document summarization is considered. This task is about extracting the general information from several texts written on the same topic. The work investigates the applicability of the variational autoencoder with Transformer architecture to the abstractive multi-document summarization. It is also proposed to use the pretrained BERT, which shows significant results in many Natural language processing tasks.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses