CS-503-A Data Analytics - Nov 2023 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 7 Marks
   aSuppose the weights of 800 male students are normally distributed with 28.8 kg and SD of 2.06 kg. Find the number of students whose weights are

   1. i)Between 28.4 kg and 30.4 kg
   2. ii)More than 31.3 kg
   How to answer

   • State mean = 28.8 kg, SD = 2.06 kg, N = 800 and normal distribution assumption
   • Convert limits to Z-scores using $Z=(X-\mu)/\sigma$
   • Find areas from normal table for i) between 28.4 and 30.4 and ii) >31.3
   • Multiply probabilities by 800 to get number of students
2. 7 Marks
   bA random variable has the following probability function: Determine:

   1. i)k
   2. ii)mean
   3. iii)variance

   

   How to answer

   • Use $\sum P(x)=1$ to form equation in K and solve quadratic, reject negative value
   • Compute mean as $E[X]=\sum xP(x)$
   • Compute $E[X^2]=\sum x^2P(x)$
   • Compute variance as $Var(X)=E[X^2]-(E[X])^2$
3. 7 Marks
   aA sales tax officer has reported that the average sales of the 500 businesses that he has to deal with during a year is Rs. 36,000 with a standard deviation of Rs. 10,000. Assuming that the sales in these businesses are normally distributed, find:

   1. i)The number of business as the sales of which are greater than Rs. 40,000.
   2. ii)The percentage of business the sales of which are likely to range between Rs. 30,000 and Rs. 40,000.
   How to answer

   • State mean Rs. 36,000, SD Rs. 10,000, N=500, normal distribution
   • Compute Z for 40,000 and 30,000
   • Find tail area >40,000 and multiply by 500 for count
   • Find area between 30,000 and 40,000 for percentage
4. 7 Marks
   bDiscuss the trends in big data generation and acquisition.
   How to answer

   • Define big data generation sources and acquisition concept
   • Discuss trends: IoT, social media, mobile, machine-generated data growth
   • Discuss acquisition trends: real-time ingestion, sensors, cloud, streaming
   • Conclude with volume-velocity-variety implications
5. 14 Marks
   Explain the following:

   1. a)Predictive analytics
   2. b)Inter-and Trans-firewall analytics
   3. c)Information management
   4. d)Crowd sourcing analytics
   How to answer

   • Define each of the four analytics terms
   • Explain purpose and key techniques for each
   • Give one example/application for each
   • Compare relevance to data analytics
6. 7 Marks
   aWith an example, explain the term social media analytics.
   How to answer

   • Define social media analytics and its need
   • Describe data sources and metrics (sentiment, engagement)
   • Give example e.g., brand monitoring / election prediction
   • Mention tools and outcome
7. 7 Marks
   bWhat are the various stages in big data analytics life cycle? Illustrate with a figure, explaining each of them.
   How to answer

   • List stages of big data analytics life cycle (discovery, ingestion, processing, analysis, visualization)
   • Draw labelled life-cycle figure
   • Explain each stage briefly
   • Mention iteration and deployment
8. 7 Marks
   aBrief about the main component of Map Reduce.
   How to answer

   • Define MapReduce paradigm
   • Describe Map, Shuffle/Sort, Reduce phases
   • Mention JobTracker/TaskTracker or main components
   • Give data flow example
9. 7 Marks
   bWhat is Hadoop? Describe the role of Hadoop in big data analysis. Also explain core components of Hadoop.
   How to answer

   • Define Hadoop and its origin
   • Explain role in storage and processing of big data
   • Describe core components: HDFS, YARN, MapReduce (and common utilities)
   • Mention scalability and fault tolerance benefits
10. 7 Marks
    aDescribe the structure of HDFS in a Hadoop ecosystem using a diagram.
    How to answer

    • Define HDFS and its design goals
    • Draw HDFS architecture with NameNode, DataNodes, blocks
    • Explain replication, read/write flow
    • Mention rack awareness and fault tolerance
11. 7 Marks
    bWhy is finding similar items important in Big Data? Illustrate using two example applications.
    How to answer

    • Define similar-items problem (near-duplicates, similarity measures)
    • Explain importance: deduplication, recommendation, search
    • Illustrate with two applications e.g., plagiarism / product recommendation
    • Mention shingling / Jaccard / LSH briefly
12. 7 Marks
    aWhy to choose Hadoop for processing Big Data in detail and explain the concept of distributed and parallel computing challenges?
    How to answer

    • List reasons: scalability, fault tolerance, commodity hardware, open source
    • Explain distributed vs parallel computing challenges
    • Relate Hadoop HDFS/MapReduce to those challenges
    • Conclude with use cases
13. 7 Marks
    bExplain in detail the interacting process with Hadoop Ecosystem. List out various big data processing technologies.
    How to answer

    • Describe Hadoop ecosystem components (Hive, Pig, HBase, Spark, etc.)
    • Explain interaction/workflow with HDFS/YARN
    • List big data processing technologies
    • Mention batch vs real-time processing
14. 7 Marks
    aExplain Pig Data Model in detail and discuss how it will help for effective data flow?
    How to answer

    • Define Pig and its data model: Atom, Tuple, Bag, Map
    • Explain Pig Latin flow and schema flexibility
    • Discuss how it simplifies ETL/data flow
    • Give example of data flow operation
15. 7 Marks
    bDraw and explain architecture of APACHE HIVE. Explain various data insertion techniques in HIVE with example.
    How to answer

    • Draw labelled APACHE HIVE architecture (driver, metastore, execution engine)
    • Explain query compilation and execution flow
    • Describe data insertion techniques: LOAD, INSERT, partitioning with examples
    • Mention HiveQL example