1️⃣ Question 1

What is the main purpose of data modeling in the data science methodology?

• ✅ To develop models for descriptive or predictive purposes

• ❌ To refine and adjust the problem statement

• ❌ To select an analytical approach

• ❌ To collect raw data

Explanation:
Modeling focuses on building descriptive or predictive models that answer the business problem.

---

2️⃣ Question 2

How does a training set contribute to predictive modeling?

• ❌ Helps select algorithms

• ✅ A training set serves as a calibration gauge for the model

• ❌ Contains variables not required

• ❌ Provides unknown outcomes

Explanation:
The training set teaches (calibrates) the model how to recognize patterns and relationships.

---

3️⃣ Question 3

Primary purpose of model evaluation?

• ✅ To assess the quality of the model and ensure it meets the initial request

• ❌ Determine parameter values

• ❌ Refine data collection

• ❌ Deploy the model

Explanation:
Evaluation verifies whether the model appropriately solves the defined business problem.

---

4️⃣ Question 4

Purpose of diagnostic measures during model evaluation?

• ❌ Refine model design

• ❌ Ensure model is functioning

• ❌ Assess descriptive relationships

• ✅ To test the model’s statistical significance

Explanation:
Diagnostic measures in evaluation check how statistically reliable and valid the model is.

---

5️⃣ Question 5

What does the ROC curve help determine?

• ❌ Statistical significance

• ❌ Optimal model based on diagnostics

• ✅ The true-positive rate and false-positive rate for different criteria

• ❌ Misclassification cost

Explanation:
The ROC curve evaluates classification performance by showing the tradeoff between TPR and FPR.

---

🧾 Summary Table