• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Fake News Detector as a New Evaluation Metric for Text GANs

Student: Bystrova Olga

Supervisor: Boris Orekhov

Faculty: Faculty of Humanities

Educational Programme: Fundamental and Computational Linguistics (Bachelor)

Year of Graduation: 2020

In this word we explore the problem of text generation and detection of generated text on news corpus. We created a dataset to deal with the task of fake news classification. It consists of real news and texts generated by GPT-2 model with different sampling types. The dataset consists of 1,2 million texts. We tried different models for the classification task. The best results were shown by LSTM and BERT models. To deal with text generation problem we used Generative Adversarial Network as a way to train pretrained language model. Different models, such as LSTM, BERT, CNN, were used as a discriminator.

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