Data Science for Engineers Week 7 NPTEL Assignment Answers 2025

NPTEL Data Science for Engineers Week 7 Assignment Answers 2025

1. Which among the following is not a type of cross-validation technique?

• a. LOOCV
• b. k-fold cross validation
• c. Validation set approach
• d. Bias variance trade off

✅ Answer :- d
✏️ Explanation: Bias-variance trade-off is a concept, not a validation technique. The other three are standard cross-validation methods.

---

2. Which among the following is a classification problem?

• a. Predicting the average rainfall in a given month.
• b. Predicting whether a patient is diagnosed with a disease or not.
• c. Predicting the price of a house.
• d. Predicting whether it will rain or not tomorrow.

Consider the following confusion matrix for the classication of Hatchback and SUV:

✅ Answer :- b, d
✏️ Explanation: Classification problems involve predicting discrete outcomes (like Yes/No, 0/1). Options b and d are binary classifications.

---

3. Find the accuracy of the model.

• a. 0.95
• b. 0.55
• c. 0.45
• d. 0.88

✅ Answer :- a
✏️ Explanation: Accuracy = (TP + TN) / Total samples. Based on the assumed values, the model achieves 95% accuracy.

---

4. Find the sensitivity of the model.

• a. 0.95
• b. 0.55
• c. 1
• d. 0.88

✅ Answer :- c
✏️ Explanation: Sensitivity = TP / (TP + FN). If there are no false negatives, sensitivity is 1, meaning the model identifies all positive cases.

---

5. Under the ‘family’ parameter of glm() function, which one of the following distributions correspond to logistic regression for a variable with binary output?

• a. Binomial
• b. Gaussian
• c. Gamma
• d. Poisson

✅ Answer :- a
✏️ Explanation: Logistic regression in R uses family = binomial for binary classification.

---

6. What is the dimension of the dataframe?

• a. (150, 5)
• b. (150, 4)
• c. (50, 5)
• d. None of the above

✅ Answer :- a
✏️ Explanation: The classic Iris dataset has 150 rows (flowers) and 5 columns (4 features + 1 label).

---

7. What can you comment on the distribution of the independent variables in the dataframe?

• a. The variables Sepal Length and Sepal Width are not normally distributed
• b. All the variables are normally distributed
• c. The variable Petal Length alone is normally distributed
• d. None of the above

✅ Answer :- b
✏️ Explanation: The Iris dataset features are often treated as normally distributed for statistical modeling.

---

8. How many rows in the dataset contain missing values?

• a. 10
• b. 5
• c. 25
• d. 0

✅ Answer :- d
✏️ Explanation: The Iris dataset is clean and contains no missing values.

---

9. Which of the following code blocks can be used to summarize the data (finding the mean of the columns PetalLength and PetalWidth), similar to the one given below?

• a. lapply(irisdata[, 3:4], mean)
• b. sapply(irisdata[, 3:4], 2, mean)
• c. apply(irisdata[, 3:4], 2, mean)
• d. apply(irisdata[, 3:4], 1, mean)

✅ Answer :- a, c
✏️ Explanation:

• lapply() and apply(..., 2, ...) apply functions column-wise.
• Option b has incorrect syntax.
• Option d calculates row-wise mean, not required here.

---

10. What can be interpreted from the plot shown below?

• a. Sepal widths of Versicolor flowers are lesser than 3 cm.
• b. Sepal lengths of Setosa flowers are lesser than 6 cm.
• c. Sepal lengths of Virginica flowers are greater than 6 cm.
• d. Sepals of Setosa flowers are relatively more wider than Versicolor flowers.

✅ Answer :- b, d
✏️ Explanation:

• Setosa has shorter sepals.
• Setosa sepals are generally wider than Versicolor, which is visible in boxplots or violin plots.