Over the past few months (from the beginning of April until the end of July), our team of data scientists took part in a large international machine learning competition that was held on the basis of Kaggle Data Science platform.
Kaggle is an on-line community of data scientists and machine learners, owned by Google LLC. It is the largest and most diverse data community in the world (the community spans 194 countries). Kaggle allows users to find and publish data sets, explore and build models in a web-based data science environment, work with other data scientists and machine learning engineers, and enter competitions to solve data science challenges.
Kaggle competitions regularly attract over a thousand teams and individuals. It was in one of these competitions that our team took place – Jigsaw Unintended Bias in Toxicity Classification. This competition was organized and held the Conversation AI team, a research initiative founded by Jigsaw and Google. The Conversation AI team builds technology to protect voices in conversations. A main area of focus is machine-learning models that can identify toxicity in on-line conversations, where toxicity is defined as anything rude, disrespectful or otherwise likely to make someone leave a discussion. That was the very aim of the competition: detect toxic comments and minimize unintended model bias.
Competing with more than three thousand teams from around the world, our team finished the competition in the medal standings and won a bronze medal. We congratulate our team on success in this competition and wish them future victories in the next competitions and challenges.