Skip to content

NLP project that detects hateful comments on youtube videos

Notifications You must be signed in to change notification settings

carrillo-p/Akrom

 
 

Repository files navigation

Akroma

Proyecto Logo

Taking it's name from the mythical angel of hate, our app understands how dangerous the internet can be. Filled to the brim with caustic comments, the experience of interacting online can often feel like navigating a minefield. Recognizing this critical issue, we are proud to unveil our revolutionary application, designed to transform the landscape of digital communication.

Leveraging cutting-edge machine learning algorithms and sophisticated neural networks, our app is a beacon of hope in the battle against online toxicity. By utilizing Natural Language Processing (NLP), it meticulously scans and identifies harmful comments in real-time, ensuring that platforms like YouTube remain safe and welcoming spaces for all users.

Our app not only detects but also mitigates toxic interactions, fostering a community built on respect and positivity. It’s more than just a tool; it’s a movement towards a healthier digital ecosystem.

Our commitment is to harness this power to protect users and promote genuine, constructive conversations online.

Join us in this endeavor to make the internet a better place. Together, we can create a safer, kinder digital world, one comment at a time.


In addition to its core functionalities, our app offers two main services tailored to user needs.

Firstly, we provide a robust testing environment. Here, you can write your own texts and experiment with our various models to determine which one best suits your specific use case. This allows you to gain a comprehensive understanding of our app's capabilities and select the most appropriate model for your requirements.

Secondly, we offer a YouTube comment analysis service. By simply introducing a YouTube link, our app can retrieve and analyze the most recent X comments, identifying and highlighting any hateful content. Moreover, for those who wish to monitor ongoing comment sections, our service can provide live updates, tracking new toxic comments as they appear on the video.


These services are designed to empower users, giving them the tools to combat online toxicity proactively and effectively. By leveraging our advanced technology, you can ensure that your digital spaces remain positive and respectful environments.

Together, let’s continue to pave the way for a more respectful and inclusive internet, one comment at a time.


Instrucciones de instalacion

Simplemente haced docker-compose, necesitais añadir vuestro propio .env en la raiz del proyecto y vuestras propias credenciales de google firebase en la carpeta database. para que funcione

About

NLP project that detects hateful comments on youtube videos

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 93.3%
  • Python 6.1%
  • CSS 0.6%