Question : In which of the following cases you can use the K-means clustering? A. Image Processing B. Customer Segmentation C. Classification of the plants D. Reducing the customer churn rate
1. A,B 2. B,C 3. A,B,C 4. B,C,D 5. A,B,C,D
Correct Answer : 5 Explanation: You can use K-Means clustering for various use cases like as below 1. In a video file you can find which all frames are most similar to each other. Attributes can be pixel brightness, color, and its location in the frame. 2. You can use the Patients attributes like sex, age, weight, height, Cholesterol, Sugar etc. to create the groups of the patients. Even you can use clustering in biology for the classification of plants and animals. 3. You can apply k-means clustering on the customers who have the similar spending patterns, and use them to increase the sales or reduce the customer churn rate.
Question : You want to apply the K-Means clustering on the total number of objects which are M, total attributes on each object are n. You need to create groups or clusters. What would be matrix dimension you would be using to store all the objects attributes?
1. MXn
2. MX5
3. nX5
4. 5X5
5. MXM
Correct Answer : 1 Explanation: As there are M objects then you can have in total M number of rows and there are n attributes then you can have n number of columns to store each object attributes. Hence, correct answer will be MXn . Number of cluster is given to create the confusion and test your knowledge.
Question : You have data of , people who make the purchasing from a specific grocery store. You also have their income detail in the data. You have created clusters using this data. But in one of the cluster you see that only 30 people are falling as below 30, 2400, 2600, 2700, 2270 etc. What would you do in this case?
1. You will be increasing number of clusters.
2. You will be decreasing the number of clusters.
3. You will remove that 30 people from dataset
4. You will be multiplying standard deviation with the 100.
Correct Answer : 2 Explanation: Decreasing the number of clusters will help in adjusting this outlier cluster to get adjusted in another cluster.
1. The change in purchase size is not practically important, and the good p-value of the second study is probably a result of the large study size. 2. The change in purchase size is small, but may aggregate up to a large increase in profits over the entire customer base. 3. Access Mostly Uused Products by 50000+ Subscribers should run another, larger study. 4. The p-value of the second study shows a statistically significant change in purchase size. The new website is an improvement.