Build Your Own AI Project Worksheet
AI tip (Accuracy = correct ÷ total × 100): Steps: 1 choose a problem → 2 collect examples → 3 label them → 4 check and correct bad labels (clean the data) → 5 split into training and test sets → 6 train → 7 test with new examples → 8 measure accuracy → 9 improve the data and retest. Accuracy = correct ÷ total × 100. Wrong = total − correct. Expected right = accuracy ÷ 100 × total. Keep some examples apart for testing (for example one fifth); the rest are for training.
NameDateTime takenScore ___ / 40
- 1.⚖️ Two recycling sorter models were tested on the same 10 waste photos. Model A got 10 right and Model B got 5 right. How many percentage points more accurate is Model A?
- 2.🎯 Meera’s spam spotter was tested on 50 new messages and got 42 right. What is its accuracy in percent?
- 3.🧪 Why must you test a model on examples it has NOT seen? (a) to make it faster (b) to save paper (c) to make it bigger (d) to see if it learned
- 4.🧪 What is accuracy? (a) total × 2 (b) correct + total (c) correct ÷ total × 100 (d) total − correct
- 5.📋 Planning an AI project: which comes first — set aside test photos or work out the accuracy? (a) set aside test photos (b) work out the accuracy
- 6.📦 Meera collected 130 labelled handwritten letters for a handwriting reader and keeps one fifth of them for testing. How many are left for training?
- 7.📋 Planning an AI project: which comes first — label each example or train the model? (a) label each example (b) train the model
- 8.🧪 Your fruit sorter mixes up oranges and lemons. What helps most? (a) a new colour screen (b) a louder speaker (c) fewer examples (d) more labelled examples
- 9.📋 Planning an AI project: which comes first — try unseen examples or set aside test photos? (a) try unseen examples (b) set aside test photos
- 10.⚖️ Two handwriting reader models were tested on the same 10 handwritten letters. Model A got 8 right and Model B got 7 right. How many percentage points more accurate is Model A?
- 11.📋 Planning an AI project: which comes first — try unseen examples or gather many examples? (a) try unseen examples (b) gather many examples
- 12.❌ Meera’s handwriting reader was tested on 21 new handwritten letters and got 20 right. How many did it get wrong?
- 13.📋 Planning an AI project: which comes first — label each example or split train and test? (a) label each example (b) split train and test
- 14.📋 Planning an AI project: which comes first — try unseen examples or train the model? (a) try unseen examples (b) train the model
- 15.🎯 Ananya’s recycling sorter was tested on 50 new waste photos and got 26 right. What is its accuracy in percent?
- 16.📋 Planning an AI project: which comes first — fix it and test again or set aside test photos? (a) fix it and test again (b) set aside test photos
- 17.📋 Planning an AI project: which comes first — improve data, retest or let it learn patterns? (a) improve data, retest (b) let it learn patterns
- 18.❌ Vihaan’s recycling sorter was tested on 40 new waste photos and got 31 right. How many did it get wrong?
- 19.📦 Ishaan collected 160 labelled leaf photos for a leaf identifier and keeps one fifth of them for testing. How many are left for training?
- 20.⚖️ Two fruit sorter models were tested on the same 25 fruit photos. Model A got 24 right and Model B got 18 right. How many percentage points more accurate is Model A?
- 21.🧪 You can plan a simple AI project with… (a) a secret password (b) only a big robot (c) nothing at all (d) paper and examples
- 22.⚖️ Two leaf identifier models were tested on the same 25 leaf photos. Model A got 23 right and Model B got 21 right. How many percentage points more accurate is Model A?
- 23.❌ Meera’s fruit sorter was tested on 21 new fruit photos and got 15 right. How many did it get wrong?
- 24.📋 Planning an AI project: which comes first — fix it and test again or test on new photos? (a) fix it and test again (b) test on new photos
- 25.📋 Planning an AI project: which comes first — let it learn patterns or try unseen examples? (a) let it learn patterns (b) try unseen examples
- 26.⚖️ Two spam spotter models were tested on the same 10 messages. Model A got 7 right and Model B got 5 right. How many percentage points more accurate is Model A?
- 27.⚖️ Two leaf identifier models were tested on the same 25 leaf photos. Model A got 21 right and Model B got 18 right. How many percentage points more accurate is Model A?
- 28.🎯 Saanvi’s handwriting reader was tested on 50 new handwritten letters and got 41 right. What is its accuracy in percent?
- 29.📦 Nisha collected 170 labelled handwritten letters for a handwriting reader and keeps one fifth of them for testing. How many are left for training?
- 30.📋 Planning an AI project: which comes first — split train and test or test on new photos? (a) split train and test (b) test on new photos
- 31.⚖️ Two leaf identifier models were tested on the same 50 leaf photos. Model A got 44 right and Model B got 43 right. How many percentage points more accurate is Model A?
- 32.❌ Meera’s leaf identifier was tested on 40 new leaf photos and got 38 right. How many did it get wrong?
- 33.📋 Planning an AI project: which comes first — train the model or fix it and test again? (a) train the model (b) fix it and test again
- 34.📋 Planning an AI project: which comes first — collect fruit photos or fix it and test again? (a) collect fruit photos (b) fix it and test again
- 35.📋 Planning an AI project: which comes first — fix it and test again or label each example? (a) fix it and test again (b) label each example
- 36.📋 Planning an AI project: which comes first — label each example or collect fruit photos? (a) label each example (b) collect fruit photos
- 37.🎯 Vihaan’s fruit sorter was tested on 25 new fruit photos and got 16 right. What is its accuracy in percent?
- 38.📋 Planning an AI project: which comes first — let it learn patterns or set aside test photos? (a) let it learn patterns (b) set aside test photos
- 39.📋 Planning an AI project: which comes first — try unseen examples or fix it and test again? (a) try unseen examples (b) fix it and test again
- 40.🎯 Kabir’s recycling sorter is 50% accurate. If it is tested on 40 new waste photos, how many should it get right?
Answer key
- 50
- 84
- to see if it learned
- correct ÷ total × 100
- set aside test photos
- 104
- label each example
- more labelled examples
- set aside test photos
- 10
- gather many examples
- 1
- label each example
- train the model
- 52
- set aside test photos
- let it learn patterns
- 9
- 128
- 24
- paper and examples
- 8
- 6
- test on new photos
- let it learn patterns
- 20
- 12
- 82
- 136
- split train and test
- 2
- 2
- train the model
- collect fruit photos
- label each example
- collect fruit photos
- 64
- set aside test photos
- try unseen examples
- 20
Free ai for students lessons, practice and worksheets at talentjr.in/ai-for-students