Jigsaw Multilingual Toxic Comment Classification

Use TPUs to identify toxicity comments across multiple languages

$50,000(Rs 37,67,625)Prize Money[to be offered by kaggle]

 

This is an event by Google and Jigsaw running on Kaggle promoted by The IoT Academy.

Register with The IoT Academy to get your internship certificate on behalf of The IoT Academy, Noida in coordination with UniConverge Technologies Pvt. Ltd. 

The IoT Academy is an NCR Based Firm founded and guided by highly experienced IT Professionals. We are a fast emerging venture imparting quality and affordable programs for colleges and individuals. We specialize in education and skill development programs for students and prepare them for a better future.

Overview

People tend to leave online conversations due to people posting toxic or disrespectful comments. You need to make a machine learning model to recognize if a comment is normal or toxic. If we can recognize such harmful contributions, we will have a healthier, more open internet.

Your training data will be the English data given for our two previous competitions(available on Kaggle) and your test data will be a commentary on the Wikipedia talk page in several languages.

Evaluation Criteria:

• Participants should submit their raw code(or give link) and accuracy report to Email id - jigsaw@theiotacademy.co for review.
• Participants should get accuracy more than 60% to get the internship certificate.
• Students should have registered through The IoT Academy link.