(SEM V) THEORY EXAMINATION 2022-23 DATA ANALYTICS

Course: B.Tech (Semester V)                                     Subject Code: KCS-051

Subject Name: Data Analytics                                 Time: 3 Hours

Total Marks: 100

Note: Attempt all sections. If any data is missing, assume suitably. 

Section A – Short Answer Questions (2 × 10 = 20 Marks)

Answer all briefly:

What are the common open-source tools for the model building phase?

What is a Decision Tree?                                          What is Learning Rate?

What is Rule Induction?                                           How can you deal with uncertainty in analytics?

What is DSMS (Data Stream Management System)?

Write names of two visualization tools.

Explain the principle behind Hierarchical Clustering.

Differentiate between Pig and SQL.

Write a short note on R graphical user interface. 

Section B – Descriptive Questions (10 × 3 = 30 Marks)

Attempt any three:

Compare Traditional Analytics Structure vs Modern Analytics Architecture.

Explain Multivariate Analysis and Bayesian Networks.

Explain the DGIM (Datar–Gionis–Indyk–Motwani) algorithm for counting ones in a window.

Why is the PCY algorithm preferred over Apriori algorithm?

How is RDBMS different from NoSQL? 

Section C – Long Answer Questions (10 × 5 = 50 Marks)

Q3

(a) Explain Apache Hadoop, KNIME, and OpenRefine in detail.
or
(b) Explain the phases of the Data Analytics Life Cycle.

Q4

(a) Differentiate between Crisp Logic and Fuzzy Logic.
or
(b) Describe Kernel Methods used in Data Analytics.

Q5

(a) Explain Bernoulli Sampling with its algorithm.
or
(b) What are the components of a general stream processing model? List some sources of streaming data.

Q6

(a) What is Prediction Error? Explain it in Classification and Regression with examples.
or
(b) Explain the SON Algorithm to find frequent itemsets using at most two passes.

Q7

(a) Draw and discuss the Architecture of Hive in detail.
or
(b) Explain approaches to integrate humans into the data exploration process to realize various types of visual data mining.