Actually, username & story titles aren't included in the model. We may experiment with including them in the future but for now we left them out to avoid any undo bias.
Sweetest users is a good question. I'll have to build out that view some time.
Your account is pretty new. We intentionally remove any accounts with less then ? number of comments to discourage people from attempting to set a record. Lol
Sorry about that, and thank you for the feedback. I'll try to make sure it is more readable in the next version. Thanks for taking the time to check it out!
The package used is a pretty popular one called TextBlob. It is nifty for working with unlabeled data like we have with the HackerNews dataset.
We really focused our definition of saltiness around being a combination of (subjective + negative) comments.
We reduced the impact of (objective + negative) as we feel that criticism, while at times painful, if presented objectively isn't necessarily salty.
We built this model fast (1 week) and have since iterated this week into developing a Fine Tuned BERT model that we are training over a much broader set of toxicity, demographic, and polarity features. The training set is much larger and higher quality so we are expecting a large jump in precision upon deployment.
I hope the app gave you some good chuckles as you went around though. It's hard to explain how excited I felt when I saw pg_is_a_butt at the top of my pandas data frame the first time I processed the data.
I agree. Would have been much easier too. Unfortunately, HN doesn't have downvotes. The current model does incorporate upvotes, but we're seeing a lot more success training with like-kind labeled datasets + BERT fine tuning.
Thank you for trying the app and hopefully V2 will leave you feeling like the system is more precise.