Current political issues are divisive and polarizing, and can create large rifts and tensions among friends, family, and colleagues. It is all too easy to unintentionally take sides and hurl harmful, unproductive comments at those you disagree with without taking the time to think about all sides of an issue. Moreover, we wanted to help users be actively conscious of whether they were engaging in inappropriate "locker room banter" to inhibit targeting of marginalized groups. We aim to provide a tool that facilitates safe and respectful discussion of these issues, as well as educate all parties involved on the content of these issues.
PCBot is a Slack bot that can be added to any Slack channel. It moderates the discussion of the channel by detecting if participants are making offensive/targeted comments towards minorities. Users will receive either private or public warnings if they post something that is deemed offensive, and the bot will also issue channel-wide warnings if there are multiple instances of inflammatory content. Moreover, in the event that a warning is issued, if the discussion centers around a currently trending political topic, the bot will post resources to educate all users on the content of the topic, encouraging each user to construct rational, factual arguments.
Tools we used: Microsoft Azure, NLP, key phrase detection, sentiment analysis, topic detection, basic feature engineering and machine learning, web scraping and HTML parsing.
Built With
- azure
- microsoft
- ml
- natural-language-processing
- python
- sentiment-analysis
- slack
Log in or sign up for Devpost to join the conversation.