Data Science for Engineers Week 8 NPTEL Assignment Answers 2025

NPTEL Data Science for Engineers Week 8 Assignment Answers 2024

1. According to the built model, the within cluster sum of squares for each cluster is _____________ (the order of values in each option could be different):

• a. 8.316061 11.952463 16.212213 19.922437
• b. 7.453059 12.158682 13.212213 21.158766
• c. 8.316061 13.952463 15.212213 19.922437
• d. None of the above

✅ Answer :- a
✏️ Explanation: These values represent the within-cluster sum of squares (WCSS), which measures the compactness of each cluster. Option (a) matches the WCSS values from the built model.

---

2. According to the built model, the size of each cluster is ______________ (the order of values in each option could be different):

• a. 13 13 7 14
• b. 11 18 25 24
• c. 8 13 16 13
• d. None of the above

✅ Answer :- c
✏️ Explanation: Cluster sizes indicate how many data points belong to each cluster. Option (c) matches the correct distribution in the model.

---

3. The Between Cluster Sum-of-Squares (BCSS) value of the built K-means model is ______________ (Choose the appropriate range)

• a. 100 – 200
• b. 200 – 300
• c. 300 – 350
• d. None of the above

✅ Answer :- a
✏️ Explanation: BCSS measures the variance between clusters. A value in the range of 100–200 is typical for a moderate model with 4 clusters.

---

4. The Total Sum-of-Squares value of the built k-means model is _________ (Choose the appropriate range)

• a. 100 – 200
• b. 200 – 300
• c. 300 – 350
• d. None of the above

✅ Answer :-
✏️ Explanation: TSS = WCSS + BCSS. Since BCSS is already in the 100–200 range, and WCSS values are moderate, TSS is also expected to lie in this range.

---

5. Which of the statement is INCORRECT about KNN algorithm?

• a. KNN works ONLY for binary classification problems
• b. If k=1, then the algorithm is simply called the nearest neighbour algorithm
• c. Number of neighbours (K) will influence classification output
• d. None of the above

✅ Answer :- a
✏️ Explanation: KNN can be used for multi-class classification as well, not just binary. Hence, statement (a) is incorrect.

---

6. K means clustering algorithm clusters the data points based on:-

• a. Dependent and independent variables
• b. The eigen values
• c. Distance between the points and a cluster centre
• d. None of the above

✅ Answer :- c
✏️ Explanation: K-means uses Euclidean distance to assign points to the nearest cluster center.

---

7. The method / metric which is NOT useful to determine the optimal number of clusters in unsupervised clustering algorithms is

• a. Scatter plot
• b. Elbow method
• c. Dendrogram
• d. None of the above

✅ Answer :- a
✏️ Explanation: A scatter plot visualizes data but does not directly help in determining the optimal number of clusters. Elbow and dendrogram methods do.

---

8. The unsupervised learning algorithm which aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest centroid is

• a. Hierarchical clustering
• b. K-means clustering
• c. KNN
• d. None of the above

✅ Answer :- b
✏️ Explanation: K-means clustering partitions data based on proximity to centroids, making it the correct answer.