Skip to main content

Your Stats

0

Total XP

0/24

Badges

0

Day Streak

0

Total Score

0 /5 levels

0 pts

Games

Data Quality

Data Cleaner

Separate clean training data from noisy, biased, or toxic examples.

+50 XP Active

Your task

Easy · 1× pts

You are auditing a training dataset. Sort each example as Clean or Noisy.

How to play: Read each item, then tap the category it belongs to.

Levels

0 pts

Label: "Cat" — Image shows a clear photo of a tabby cat.

Label: "Dog" — Image is completely black with no visible subject.

Label: "Spam" — Email text: "CONGRATULATIONS!!! You won a free iPhone!!!"

Label: "Positive Review" — Text: "asdfgh jkl 12345 random"

Label: "Stop Sign" — Image shows a red octagonal stop sign on a clear road.

Label: "Healthy" — Medical record is completely empty.

0/6 sorted
What this means

"Garbage in, garbage out" is the golden rule of AI. A model trained on bad data produces bad results — no matter how clever the algorithm. Data cleaning means finding and removing duplicates, errors, bias, and noise before training. It’s the least glamorous and most important part of AI.

Leaderboard

Loading…