How I Built an Intelligent Twitter Bot
Building a Twitter Bot to Combat Spam
Twitter users tweet 500 million tweets per day. The volume of information going through Twitter per day makes it one of the best platforms to get information on any subject of interest. In this post, I’ll walk you through how I built a twitter bot with a brain — powered by machine learning.
According to a report, there are 48 million bots on twitter — and hubofml happened to be one of those bots. Hubofml started as a simple bot written in Node.Js (on one Sunday evening) running on a free Heroku dyno that tracks specific hashtags like machinelearning, computervision, and retweet tweets containing those hashtags.
My goal was to use the bot to collate information on machine learning and re-broadcast to people interested in them — after all, some of the best posts I’ve read on machine learning came from links shared on Twitter by the community.
I thought it would be cool to have a bot that tracks hashtags related to machine learning that I follow to stay informed.
Days after the bot was deployed, I started noticing some forms of abuse and spam like this:
