• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site
For visually-impairedUser profile (HSE staff only)SearchMenu

Analysis of Genre Dependencies of Visual Elements of Books Using Deep Learning Methods

Student: Asia Kupinskaia

Supervisor:

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

Educational Programme: Big Data Analysis for Business, Economy, and Society (Master)

Year of Graduation: 2021

Book covers give a first impression of the book's content, subject matter and central idea. However, the classification of book covers is a difficult task due to its subjectivity and vagueness of class. Within the framework of this work, the task is to analyze the effectiveness of modern deep learning models for identifying book genres based only on its cover. Performance is measured on a number of modern convolutional neural network models. The attention mechanism is then used to identify the regions the network is focused on in order to make a prediction. Analysis of the results shows that the current generation of models cannot solve this problem with a satisfactory level of efficiency.

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