This neural network can help determine the correspondence of the attached video topic to the video topics recommended by YouTube.
Before running repositorie please visit https://chromedriver.chromium.org and make you sure you are using correct version of chromedriver. If you still have questions, please write to me.
In this project, I used a smart neural network model SBERT to calculate the proximity between the topic entered by the user and the titles of videos recommended on YouTube.
All libraries for this project are described in requests.txt. The project itself is located in sbert.py file.
Parsing is also used in this project. For him, I installed the selenium libraries and beautiful soup 4. When installing chromedriver on the server, unfortunately, an error occurs that I could not solve, so a video of the project's work will be attached below. You can also run the project yourself, since the error is observed only on the server, and not on the local computer.
This is my first pet project in data science.
Thanks for your attention!