Analysis of Affine Transformations in Semantic Spaces of Contextualisation Models

Student: Kulichenko Vladimir

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

Educational Programme: Applied Mathematics (Bachelor)

Year of Graduation: 2021

Word embedding models are able to solve various types of NLP tasks. The ability of an embedding model to solve the proportional analogy task is a popular metric of its quality. In this thesis, the proportional analogy task will be solved using affine transformation of the parallel translation in semantic spaces of word vectors, embedded using pretrained BERT and ELMo models. The results of numerical experiments revealed, that this method of solving proportional analogy problem does not always gives a correct result.

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