Skip to content

Latest commit

 

History

History
9 lines (6 loc) · 910 Bytes

README.md

File metadata and controls

9 lines (6 loc) · 910 Bytes
Raw

Deep Learning for Text Analysis

The purpose of this research is to investigate the problems of plagiarism and artificial generation of text content, ways to solve them, and create a solution for identifying these problems in texts.

  • The process and results of comparing text embedding models for further fine-tuning are presented in notebook paraphrasing/Paraphrasing_graph_comparison.ipynb.
  • The results of training and fine-tuning the final model to solve a problem of identifying artificial text generation is presented in a notebook en_generating/en_generating_bert.ipynb.
  • The results of training and fine-tuning the final model to solve a problem of artificial text in notebook generationparaphrasing/paraphrasing_bert.ipynb.
  • The notebook paraphrasing/Paraphrasing_graph_final.ipynb provides a detailed analysis of the trained final model for the task of identifying paraphrasing in the text.