CLASS- X (AI)

EVALUATION WORKSHEET

Question Choice 1 Choice 2 Choice 3 Choice 4

What is the main goal of To improve the To understand To increase model To reduce data
evaluation in AI models? training dataset model reliability complexity usage
What phenomenon
occurs when a model
Underfitting Overfitting Generalization Data leakage
remembers the training
set?
What term describes
when both prediction and True Positive True Negative False Positive False Negative
reality indicate no fire?
What is a False Positive Fire predicted
Fire predicted No fire predicted No fire predicted
in the context of forest when there is
correctly when there is one correctly
fire prediction? none
Which matrix helps in
Performance
mapping prediction Decision matrix Confusion matrix Accuracy matrix
matrix
against reality?
What does accuracy Total Percentage of
Number of training Complexity of the
measure in model predictions correct
samples model
evaluation? made predictions
What is a False Negative
Fire predicted No fire predicted No fire predicted Fire predicted
in the forest fire
correctly correctly when there is one when there is none
scenario?
Why shouldn't training
It leads to It increases It improves model It reduces
data be used for
overfitting accuracy training computation time
evaluation?
Correct
Correct Incorrect Incorrect prediction
What is a True Positive? prediction of no
prediction of fire prediction of fire of no fire
fire
Mapping of
What does the confusion Evaluation Understanding of
Prediction results predictions and
matrix NOT provide? metrics model performance
reality

Question Choice 1 Choice 2 Choice 3 Choice 4

Correctly Incorrectly Correctly Incorrectly
What does True Positive
predicted positive predicted positive predicted predicted
(TP) indicate?
cases cases negative cases negative cases
Correctly Incorrectly Correctly Incorrectly
What is True Negative
predicted predicted negative predicted predicted
(TN)?
negative cases cases positive cases positive cases
What does a high More False Fewer False More False More True
Precision value imply? Positives Alarms Negatives Negatives
Which of the following
False Negative
increases if there are Precision Recall Accuracy
rate
many False Positives?
 Fraction of actual Fraction of actual Total number of Total number of
What does Recall
positives correctly negatives correctly correct incorrect
measure?
identified identified predictions predictions
In the context of fire Fire detected
Fire not detected Correct Correct detection
detection, what is a False when there is
when there is one detection of fire of no fire
Negative? none
What happens if Precision More false alarms Fewer false alarms Increased True Increased True
is low? occur occur Positives Negatives
Which metric does not
consider False Precision Recall Accuracy F1 Score
Negatives?
What is the consequence
No False No False More True More False
of a model with 100%
Positives Negatives Negatives Alarms
Precision?
If a model has high It correctly
It misses many It has many It has low
Recall, what does it identifies most
positive cases false alarms accuracy
signify? positive cases

Question Choice 1 Choice 2 Choice 3 Choice 4

What do Precision and
Recall both share in their True Positives False Positives False Negatives True Negatives
numerator?
What does Precision
False Positives True Positives True Negatives False Negatives
measure in its denominator?
What does Recall consider in
True Positives False Negatives False Positives True Negatives
its denominator?
To balance
What is the purpose of the To measure To count True To evaluate
Precision and
F1 Score? accuracy Positives model speed
Recall
When both
When is the F1 Score When When Recall is When F1 Score
Precision and
considered perfect? Precision is 0 0 is 0.5
Recall are 1
What range does the F1
0 to 10 0 to 1 0 to 100 0 to 50
Score fall within?
If a model has high Precision The model is The model has The model is
The model is
but low Recall, what can be performing balanced biased towards
performing well
said? poorly performance recall
What happens to the F1
It remains It becomes
Score if both Precision and It increases It decreases
unchanged perfect
Recall are low?
Which scenario indicates a High Precision, Low Precision,
High F1 Score Low F1 Score
good model performance? Low Recall High Recall
Which measure is NOT
directly considered in the F1 Precision Recall True Positives False Negatives
Score calculation?