Machine learning analysis of network traffic
Name:
Date:
Class:
A
Quick Questions
Shade in the A or B circle for each question.
1. Machine learning analysis can detect traffic patterns that don't match fixed, predefined rules.
2. A machine learning model must be trained before it can classify network traffic accurately.
3. Which classifies traffic in ML-based analysis: a trained model, or a fixed list of rules?
4. ML-based analysis can only flag threats identical to ones it has seen before.
5. Which needs prior training on data: a machine learning model, or a fixed rule-based system?
B
Match the Pairs
Draw a line connecting each item on the left to its match on the right.
| Left column | Right column | |
|---|---|---|
| 1pred = classify(model, traffic) | ALoading a previously trained model for use | |
| 2model = load_model() | Bif pred == "anomaly": alert() | |
| 3ML models learn patterns from training data rather than fixed rules | Cmodel = train_model(data) | |
| 4Classifying traffic flags it as normal or anomalous based on learned patterns | DThe mistake: swapping the argument order for classify() | |
| 5pred = "normal" | EThe mistake: hardcoding the prediction instead of using the model | |
| 6pred = classify(t, model) | FClassifying traffic using a trained model |
C
Definitions
Machine learning traffic analysis
D
Fill in the Blanks
Complete each line of code by filling in the missing parts.
1.
_____ = load_model()
2.
traffic = ___________()
3.
prediction = ________(model, traffic)
4.
if prediction __ "anomaly":
5.
def ______________(model, traffic):
E
Spot the Error
Shade in the letter of the line that contains an error.
Question 1
A
model = None
B
t = get_traffic()
C
pred = classify(model, t)
Question 2
A
skip_model = True
B
model = load_model()
C
pred = classify(model, t)
Question 3
A
model = load_model()
B
t = None
C
pred = classify(model, t)
Question 4
A
model = load_model()
B
use_fixed_rules = True
C
t = get_traffic()
Question 5
A
model = load_model()
B
t = get_traffic()
C
pred = "normal"
Answer Key
Machine learning analysis of network traffic
Section A: Quick Questions
Q1
B — True
Q2
B — True
Q3
A — A trained model
Q4
A — False
Q5
A — A machine learning model
Section B: Match the Pairs
Q1
pred = classify(model, traffic) → Classifying traffic using a trained model
Q2
model = load_model() → Loading a previously trained model for use
Q3
ML models learn patterns from training data rather than fixed rules → model = train_model(data)
Q4
Classifying traffic flags it as normal or anomalous based on learned patterns → if pred == "anomaly": alert()
Q5
pred = "normal" → The mistake: hardcoding the prediction instead of using the model
Q6
pred = classify(t, model) → The mistake: swapping the argument order for classify()
Section C: Definitions
Q1
Machine learning traffic analysis — Using a trained model to classify network traffic as normal or anomalous, based on learned patterns rather than fixed rules.Section D: Fill in the Blanks
Q1
model = load_model()Q2
traffic = get_traffic()Q3
prediction = classify(model, traffic)Q4
if prediction == "anomaly":Q5
def classify_batch(model, traffic):Section E: Spot the Error
Q1
A — Line 1 sets model to None instead of a trained model, so there's nothing for classify() to use.
Q2
A — Line 1 sets a flag to skip loading the model, but classify() still needs a trained model to run.
Q3
B — Line 2 sets t to None instead of real traffic data, so there's nothing for the model to classify.
Q4
B — Line 2 switches to fixed rules instead of the trained model, defeating the point of ML-based analysis.
Q5
C — Line 3 hardcodes the prediction as normal instead of calling classify() to analyse the traffic.