I added a flag for technical and then sort it by hot [1]. I just needed to train a classifier so there's less manual classification needed. Maybe something lik

I added a flag for technical and then sort it by hot [1]. I just needed to train a classifier so there's less manual classification needed. Maybe something like a naive bayes [0] will work? I'm not sure if it's the best dataset. I could make my own dataset by just looking at how many removed comments there are :P I haven't implemented a classifier before so it may take me a few days. If you'd like to be notified when it's actually working send me an email at info@{the site name} [0] - https://www.kaggle.com/kinguistics/classifying-news-headlines-with-scikit-learn https://www.kaggle.com/kinguistics/classifying-news-headline... [1] - https://efficientdemocracy.com/technical-and-hot https://efficientdemocracy.com/technical-and-hot