Context-Sensitive Paraphrasing with Deep Learning

Student: Permyakov Artyom

Supervisor: Ivan Bliznets

Faculty: St. Petersburg School of Physics, Mathematics, and Computer Science

Educational Programme: Enterprise Software Development (Master)

Year of Graduation: 2021

The majority of previous research on paraphrase generation has focused on paraphrasing the entire text. However, in practice this is not always necessary. This can happen, for example, the name of the capital of one state is replaced by the name of the capital of another state during paraphrasing. In some situations, such a substitution may not be desirable. The model, which is able to partially paraphrase the text, can be used as an assistant when writing letters, books, academic articles. Also, this functionality can be useful in search and question-answering systems. Due to the practical benefits of partial paraphrasing and the absence of an explicit solution for this problem, the creation of such a model is relevant. This work aims to create a model for partial paraphrasing.

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