DAYANANDASAGAR COLLEGE OF ENGINEERING

DEPARTMENT OF MCA

Question Bank

Semester: IV Sem Subject: Data Science Sub Code: 17MCA441

Module 1

Sl.No. Questions Marks

1 What is Data Mining? Explain the process of knowledge discovery in data 8marks
bases
2 Explain the motivating challenges in data mining 8 marks
3 Describe data mining as a confluence of many disciplines 8 marks
4 What are data mining tasks? Explain 8 marks
5 Explain the market basket analysis in data mining 8 marks
6 Describe the application areas in data mining 8 marks
7 What is classification? Explain 10marks
8 List out some applications in classification 10marks
9 What is clustering? List out some applications in clustering 10marks
10 10marks

Module 2

Sl.No. Questions Marks

1 What is Data? What are the types of Data 8marks
2 Define an attribute? Explain types of attributes 8 marks
3 Mention the general characteristics of data sets 5marks
4 Discuss specific aspects of data quality 8 marks
5 Mention the data quality issues in mining the data 8 marks
6 What is data preprocessing? Explain some methods for preprocessing the 8 marks
data.
7 What is dimensionality reduction? Explain the curse of dimensionality 10marks
8 Explain the approaches in feature subset selection 10marks
9 Explain discretization and binarization in data mining 10marks
10 What are the measures of similarity and dissimilarity 10marks
11 Explain the methods to measure the similarities between the data objects 8 marks
12 What are the methods to measure the dissimilarities between the data 8 marks
objects
13 What is correlation ?Explain 8 marks
14 What are the issues in proximity calculation 8 marks
15 Describe the general characteristics of selecting the right proximity 8 marks
 measure
16 Distinguish between noise and outliers 8 marks
17 Which approach jaccard or hamming distance is more similar to simple 8 marks
matching coefficient
18 Which approach is more similar to the cosine measure 8 marks
19 For the following vectors,x and y calculate the indicated similarity or 8 marks
distance measure
X=(1,1,1,1)Y=(2,2,2,2) cosine,correlation ,Euclidean
X=(0,1,0,1)Y=(1,0,1,0) cosine,correlation,Euclidian,jaccard
20 Explain why computing the proximity between two attributes is often 8 marks
simpler than computing the similarity between two objects.

Module 3

Sl.No. Questions Marks

1 Why use support and confidence? 5 marks
2 What is association rule mining? Explain 8 marks
3 How to generate the frequent item sets? 8 marks
4 Write the steps to generate the frequent item sets using Aprioiri algorithm 10 marks
5 Explain about the candidate generation and pruning. 10 marks
6 Describe the support counting using a hash tree 10 marks
7 How to measure the computational complexity in data mining 8 marks
8 How to generate the rules in apriori algorithm 8 marks
9 Describe the maximal frequent item sets 5 marks
10 Explain the closed frequent item sets 5 marks
11 Describe all the alternative methods for generating frequent item sets. 10 marks
12 Construct the FP-tree algorithm with an example. 10 marks
13 How to generate the frequent item set generation in FP-growth algorithm 10 marks
14 How to evaluate the association patterns 8 marks
15 What are the objective measures of interestingness 5 marks
16 Write the limitation of interest factor, correlation analysis, and IS 8 marks
measures.
17 How to discover association rule using hash tree explain 10 marks
18 Mention the factors affecting the complexity 5 marks
19 Differentiate between the maximal and closed frequent item sets. 5 marks
20 How to compute the interestingness measures. 5 marks
21 What is the effect of support based prunining 5 marks
22 What is statistical independence and statistical based measures 5 marks
23 Distinguish between interestingness and unexpectedness 5 marks
Module 4

Sl.No. Questions Marks

1 Describe the general approach to solving a classification problem 8 marks
2 How a decision tree works. 8 marks
3 How to build a decision tree. 8 marks
4 Describe how a decision tree grows recursively using Hunt’s Algorithm. 10 marks
5 Describe the design issues of decision tree induction. 8 marks
6 Explain the methods for expressing attribute test conditions. 8 marks
7 What are the measures for selecting the best split? 8 marks
8 Explain how to split the binary, nominal and continuous attributes. 8 marks
9 Write an algorithm for decision tree induction. 8 marks
10 Explain rule based classifier? 8 marks
11 How a rule based classifier works? 8 marks
12 What are the rules ordering schemes? 8 marks
13 How to build a rule based classifier? 8 marks
14 Explain the RIPPER algorithm used for rule induction. 8 marks
15 Describe indirect methods for rule extraction. 8 marks
16 Mention the characteristics of rule based classifiers. 8 marks
17 Describe the nearest neighbor algorithm. 8 marks
18 Write the k-nearest neighbor classification algorithm. 8 marks
19 What are the characteristics of nearest neighbor classifiers? 8 marks
20 Describe the Bayesian classifiers for classification. 8 marks
21 Explain the naïve Bayes classifier. 8 marks
22 How a naïve Bayes classifier works. 8 marks
23 How to estimate conditional probabilities for categorical attributes 8 marks
24 How to estimate conditional probabilities for continuous attributes 8 marks
25 Describe the M-estimate of conditional probability. 8 marks
26 What are the characteristics of naïve Bayes classifiers? 8 marks
27 How to measure the Bayes error rate. 8 marks
28 What are Bayesian Belief networks? Explain 8 marks

Module 5

1 What is cluster Analysis? List out the application areas of cluster analysis 10 marks
to practical problems.
2 Explain the different types of clustering. 8 marks
3 Describe the basic K-means algorithm with example. 8 marks
4 Mention the ways in choosing the initial centroids. 8 marks
5 Determine the time and space complexity of K-means algorithm. 5 marks
6 Explain Bisecting K-means algorithm. 8 marks
7 Mention the strengths and weakness of K-means algorithm 5 marks
8 Describe the agglomerative hierarchical clustering algorithm. 8 marks
9 Write a basic agglomerative hierarchical clustering algorithm. 8 marks
10 Explain the different ways in defining the proximity between clusters. 8 marks
11 Determine the time and space complexity of hierarchical clustering 8 marks
algorithm.
12 Illustrate Ward’s method in finding the proximity between two clusters. 8 marks
13 Describe DBSCAN clustering algorithm with example. 8 marks
14 How to evaluate clusters? Explain 8 marks
15 What is anomaly detection? Illustrate applications for which anomalies are 8 marks
of interest.
16 Mention the causes for anomalies. 8 marks
17 Explain different approaches to anomaly detection. 8 marks
18 Explain the different issues that need to be addressed when dealing with 8 marks
anomaly detection.
19 Describe the statistical approaches to outlier detection. 8 marks
20 Explain the proximity based outlier detection. 8 marks
21 Explain the clustering based techniques for outlier detection. 8 marks