halfak_wmf

4 ms·

1. 1.▲
   by halfak_wmf 10y ago

   Hey folks. I just put this dataset together. It uses a prediction model that we've been working on for a while. This helps address some of the propensity modeling issues that past researchers have looked at. This dataset should mak
2. 2.▲
   New dataset shows fifteen years of Wikipedia’s quality trends(blog.wikimedia.org)

   3 points by halfak_wmf 10y ago|2 comments
3. 3.▲
   by halfak_wmf 11y ago

   Actually our AUC is 90% and that beats the state of the art in the literature. Accuracy is easy to game. We could get 97.5% accuracy by just predicting False all of the time. We don't advise reviewers to use the prediction directly,
4. 4.▲
   by halfak_wmf 11y ago

   Sorry about the formatting issues. I can't seem to edit the post :/
5. 5.▲
   by halfak_wmf 11y ago

   +1 for Egalitarianism. So, when I use the word "feminism", I'm referring to the academic field that explores power structures in social systems. These power structures go far beyond gender issues despite what the label &quo
6. 6.▲
   by halfak_wmf 11y ago

   Actually, we openly report a wide variety of statistics about our models. E.g. if you go to https://ores.wmflabs.org/scores/enwiki , the service lists out the models that are available for English Wikipedia along with
7. 7.▲
   by halfak_wmf 11y ago

   Yeah. We're actively working with patrollers to chase down mistakes and retrain the models. This is a system that will require continual improvement. Luckily, there's a lot of researchers working on this problem so we're d