Previous-year questions, by module
Every question from the six Mumbai University papers (May 2023 to May 2026), sorted by syllabus topic. Repeats are merged into one entry; other wordings sit under “Also asked as”.
How the paper works
- 3 hours, 80 marks. Same format in all six papers.
- Q1 is compulsory: four short questions (a to d), all to be answered.
- Q2 to Q6: answer any 3. Each has two parts (a, b).
- Most papers mix in 2 to 4 numericals: Flajolet-Martin, Bloom filter, MapReduce word count, Girvan-Newman or R scripts.
- Dec 25 had two corrections from the university (Q3a, Q5a). The corrected versions are used here.
Times asked, by module
Asked in 3 or more of the 6 papers
- 5×M1 · 1.1Mention four characteristics of big data and explain them in detail.
- 5×M4 · 4.4Show how the Flajolet-Martin algorithm estimates the number of distinct elements in the stream, using the hash function given.
- 4×M2 · 2.3 · 2.4Write MapReduce pseudo code for the word count problem. Apply it to the document below and show all the steps.
- 4×M5 · 5.3Determine the communities in the given social network graph using the Girvan-Newman algorithm.
- 3×M1 · 1.4Describe the components of the Hadoop ecosystem with the help of a diagram.
- 3×M3 · 3.1List and explain the core business drivers behind the NoSQL movement.
- 3×M4 · 4.1With a neat sketch, explain the architecture of a data-stream management system.
Introduction to Big Data and Hadoop
1.1Big Data characteristics and types
-
5×Mention four characteristics of big data and explain them in detail.Also asked as
- Mention four characteristics of big data. Elaborate these characteristics with respect to social media websites. May 23, Dec 25
- Describe the characteristics of big data with an example. May 26
1.2 · 1.3Traditional vs Big Data business approach, case studies
-
1×How does big data analysis provide a competitive advantage to industries?
1.4Hadoop, core components and ecosystem
-
3×Describe the components of the Hadoop ecosystem with the help of a diagram.Also asked as
- Draw a diagram of the typical Hadoop ecosystem and explain any two of its components. Dec 24
- Explain the Hadoop ecosystem in detail. Dec 25
-
1×Explain how big data problems are handled by the Hadoop system.
Hadoop HDFS and MapReduce
2.1 · 2.2Distributed file systems and HDFS
-
1×Describe HDFS. How does HDFS manage the storage and replication of data across a distributed cluster of machines?
-
1×Distinguish between the NameNode and the DataNode.
-
1דThe Secondary NameNode is a backup of the NameNode.” Is this statement true or false? Justify your answer.
2.3 · 2.4Map tasks, grouping by key, reduce tasks, combiners, execution
-
4×Write MapReduce pseudo code for the word count problem. Apply it to the document below and show all the steps.Dec 24This is NoSQL. NoSQL handles complex data.May 25 · Dec 25Big data is powerful. Big data drives decisions.Also asked as
- May 24 gave no document: “Illustrate with an example showing all the steps.” May 24
-
2×Explain the MapReduce execution pipeline with a suitable example.Also asked as
- Explain the Shuffle & Sort phase and the Reducer phase in MapReduce. May 25
-
1×What is the role of a combiner in the MapReduce framework? Explain with an example.
2.6Matrix-vector multiplication by MapReduce
-
1×Write MapReduce pseudo code for matrix-vector multiplication. Show all the steps for the matrix and vector below.M =123456789×v =123
-
1×Write MapReduce pseudo code to multiply two matrices. Apply MapReduce to the multiplication below.A =1234×B =6789
Matrix-matrix multiplication is not named in the syllabus (2.6 says matrix-vector), but it is in the same textbook section (MMDS 2.3.9).
2.7Relational-algebra operations by MapReduce
-
2×Explain the selection and projection relational-algebra operations using MapReduce.Also asked as
- Explain the MapReduce task and the following operations using MapReduce: 1) selection, 2) projection. May 26
-
2×Explain the natural join and the grouping and aggregation relational-algebra operations using MapReduce.
Natural join and grouping/aggregation are not listed in 2.7, which stops at union, intersection and difference. They are in the same textbook section (MMDS 2.3).
2.8Hadoop limitations
-
1×Write down the limitations of Hadoop.
2.5 coping with node failures was never asked on its own.
NoSQL
3.1Introduction to NoSQL, business drivers
-
3×List and explain the core business drivers behind the NoSQL movement.
-
1×Explain CAP. How is CAP different from the ACID property in databases?
CAP is not named in the syllabus. It fits 3.1 (why NoSQL) and 3.4 (distribution models).
3.2NoSQL data architecture patterns
-
1×List the architectural patterns in NoSQL databases. Discuss the key-value and document-oriented patterns, focusing on their characteristics, use cases and examples.
-
1×Explain the following NoSQL data architecture patterns in detail: 1) key-value stores, 2) graph stores.
-
1×What is a key-value store? What are the benefits of using a key-value store?
-
1×What is a graph store? Give an example where a graph store can be used to effectively solve a particular business problem.
-
1×Write a short note on variations of NoSQL architectural patterns.
3.3NoSQL case study
-
2×Summarize how business problems have been solved faster, cheaper and more effectively using NoSQL, considering Google’s Bigtable case study. Also illustrate the business drivers and findings in it.
3.4NoSQL solutions for big data
-
1×Explain the term NoSQL. Differentiate between master-slave and peer-to-peer distribution models.
Mining Data Streams
4.1The stream data model, DSMS, stream queries, issues
-
3×With a neat sketch, explain the architecture of a data-stream management system.Also asked as
- What is a data stream management system? How does it work? May 26
-
1×List and explain the different issues and challenges in data stream query processing.
4.3Filtering streams: Bloom filter
-
1×Explain the concept of a Bloom filter with an example.
-
1×Create a Bloom filter with the following parameters. Bit array size m = 8. Hash functions: h1(x) = x mod m, h2(x) = (2x + 1) mod m, h3(x) = (3x + 2) mod m.
- Insert the elements 12, 25, 30 and 5 into the Bloom filter.
- Check whether the elements 6 and 55 are present in the Bloom filter.
- Discuss the results of your checks: which element is a true positive and which is a true negative.
-
1×Discuss the Bloom filter in detail and insert the elements 10 and 7 into a Bloom filter of size 5 using the hash functions h1(x) = x mod 5 and h2(x) = (2x + 6) mod 5.
This is the corrected version the university sent on 14 Nov 2025. The scanned paper printed the hash functions with stray “1-” and “2-” labels.
-
1×What is a false positive in a Bloom filter? Give an example.
-
1×Why is recall 100% for a Bloom filter? Explain in detail.
4.4Counting distinct elements: Flajolet-Martin
-
5×Show how the Flajolet-Martin algorithm estimates the number of distinct elements in the stream, using the hash function given.
Paper Stream Hash function Treat result as May 23 2, 1, 6, 1, 5, 9, 2, 3, 5 h(x) = 4x + 1 mod 16 4-bit binary May 24 4, 2, 5, 9, 1, 6, 3, 7 h(x) = 3x + 7 mod 32 5-bit binary Dec 24 9, 8, 7, 6, 5, 4, 3, 2 h(x) = (2x + 1) mod 32 5-bit binary May 25 10, 12, 8, 15, 6, 9, 14, 7 h(x) = 5x + 11 mod 32 5-bit binary Dec 25 3, 1, 4, 1, 5, 9, 2, 6, 5 h(x) = 3x + 7 mod 8 5-bit binary* * Dec 25 as corrected by the university (14 Nov 2025): mod 8 gives values 0 to 7, which need only 3 bits, yet the paper says “5-bit binary integer”. Copied as printed. May 23 also says “h(x) = ax + b mod 16 for some a and b” before giving 4x + 1.
-
1×Which algorithm can be used to find distinct elements in a long data stream? Give two real-life applications of it.
4.5Counting ones in a window: DGIM
-
2×What is the DGIM algorithm? Explain it with an example.Also asked as
- Explain the DGIM algorithm for counting ones in a stream, with an example. May 24
-
2×List all six constraints that must be satisfied when representing a stream by buckets in the DGIM algorithm, with examples.Also asked as
- List any five constraints that must be satisfied when representing a stream by buckets using DGIM, with examples. Dec 24
4.2 sampling data in a stream and decaying windows were never asked.
Real-Time Big Data Models
5.1Recommendation systems: content-based and collaborative filtering
-
2×Define collaborative filtering. Using an example of an e-commerce site like Flipkart or Amazon, describe how it can be used to recommend products to users.Also asked as
- Describe collaborative filtering in a recommendation system. May 24
-
1×Predict the rating of User 2 for Movie 3 using Pearson’s correlation coefficient.
Users Ratings Movie 1 Movie 2 Movie 3 Movie 4 Movie 5 User 1 2 3 2 1 1 User 2 3 4 ? 4 3 User 3 4 3 4 3 3 -
2×Explain content-based filtering with an example.Also asked as
- Justify the use of a content-based recommendation system with a specific case study. Dec 24
-
1×Describe the terms TF and IDF with respect to content-based filtering.
-
1×What is a recommendation system? How is a classification algorithm used in a recommendation system?
-
1×Define distance measure and give the formula of cosine similarity.
Distance measures are not named in the syllabus. They come from MMDS ch. 3 and are used for similarity in 5.1.
-
1×Explain Euclidean distance and Jaccard distance with an example.
5.3Social networks as graphs, communities
-
4×Determine the communities in the given social network graph using the Girvan-Newman algorithm.May 23May 24May 25 · Dec 25 (same graph)Also asked as
- Demonstrate the different communities using the Girvan-Newman algorithm for the following graph. Dec 25
-
1×Write an algorithm for the Clique Percolation Method and discover the communities in the graph below using CPM with clique size k = 3.
The paper’s graph has no node labels. The numbers 1 to 8 are added here so you can name cliques; the edges are as printed.
-
1×Write a note on social networks as graphs.
5.2 product recommendation case study was never asked by name.
Data Analytics with R
6.1Basic features of R: vectors, objects, data structures
-
2×List and discuss the various types of data structures in R.
-
1×The table shows the number of units of different products sold on different days. Create five sample numeric vectors from this data.
Product Monday Tuesday Wednesday Thursday Friday Bread 12 3 5 11 9 Milk 21 27 18 20 15 Cola Cans 10 1 33 6 12 Chocolate bars 6 7 4 13 12 Detergent 5 8 12 20 23 Asked together with “Name and explain the operators used to form data subsets in R” (6.3, below).
6.2Reading datasets and exporting data
-
1×Write a script to create a dataset named data1 in R containing the following text: 2, 3, 4, 5, 6.7, 7, 8.1, 9.
6.3Manipulating and processing data in R
-
2×Using the data frame below, create subsets and demonstrate the output.
- Create a subset of course less than 3 (May 23) / less than 5 (May 24) using [ ] brackets.
- Create a subset where course is less than 3 or class equals 2 (May 23) / course is less than 4 or class equals 1 (May 24) using the subset() function.
course id class marks 1 11 1 56 2 12 2 75 3 13 1 48 4 14 2 69 5 15 1 84 6 16 2 53 -
2×Create a data frame in R for the project data below and display the output.
- Dec 24 (ABC Corp, Mr. Thomas): show the structure and summary statistics of the data frame.
- May 25 (XYZ Ltd., Ms. Meera): two new projects were approved. Update the data frame to include them and show the final output.
Dec 24ProjectId ProjectName Budget 1 Website Redesign 150000 2 Mobile App Launch 100000 3 Data Migration 80000 4 AI Development 200000 5 Cybersecurity Audit 50000 May 25Project Id Project Name Budget Status 1 CRM Implementation 120000 In Progress 2 Cloud Infrastructure 180000 Completed 3 Network Upgrade 60000 Not Started 4 E-Commerce Platform 220000 Completed 5 Data Analytics 90000 In Progress May 25 · new projects to addProject Id Project Name Budget Status 6 UX Research 160000 Not Started 7 Cloud Integration 190000 Not Started -
2×List and explain the functions provided by R to combine different sets of data.
-
1×Name and explain the operators used to form data subsets in R.
6.4Functions and built-in functions
-
1×Write a script to sort the values in the vector (46, 23, 15, 38, 98, 56, 28, 78) in ascending and descending order. Demonstrate the output.
6.5Data visualization
-
2×Explain the different data visualization techniques in R with examples.Also asked as
- Describe the applications of data visualization. May 24