Paraphrase Identification on Russian

Student: Bakhanova Mariia

Supervisor: Ekaterina Artemova

Faculty: HSE Tikhonov Moscow Institute of Electronics and Mathematics (MIEM HSE)

Educational Programme: Applied Mathematics (Bachelor)

Year of Graduation: 2019

Paraphrase identification is the task of examining whether two sentences have the same meaning. A number of neural models were proposed and examined on English sentences. This study explores existing methods for paraphrase detection and compares their performance on Russian dataset of sentences ParaPhraser. In this paper, two methods are considered: Word2Vec with TF-IDF weighting and Bilateral Multi-perspective Matсhing. The first method estimates semantic similarity of sentences based on word embeddings whereas the second one regards dependences in context and matches units of two sentences while training. The main objective of the research is to evaluate and compare the accuracy of chosen models applied to the problem of paraphrase identification in Russian.

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