Papers / AKTU / B.Tech / Data Analytics
AKTU · B.Tech · 5th semester

AKTU Data Analytics (BCS052) Question Paper 2025-26

UniversityDr. A.P.J. Abdul Kalam Technical University (AKTU), Lucknow
CourseB.Tech, Semester V
SubjectData Analytics (BCS052)
ExamTheory Examination 2025-26 (odd semester, held January 2026). The paper is printed in English and Hindi.
Time · Marks3 hours · 70 marks
SourceAKTU question paper, via Notes Gallery
View original PDF See important topics

Note: Attempt all sections. In case of any missing data, choose suitably.

Section A

Q1. Attempt all questions in brief · 2 × 7 = 14 marks

  1. What is Data Analytics? [CO1]
  2. List any two characteristics of Big Data. [CO1]
  3. What is regression analysis? [CO2]
  4. What is a data stream? [CO3]
  5. Define frequent itemset. [CO4]
  6. What is Hadoop? [CO4]
  7. What is Exploratory Data Analysis (EDA)? [CO5]

Section B

Q2. Attempt any three · 7 × 3 = 21 marks

  1. Explain the phases of the Data Analytics Lifecycle with objectives of each phase. [CO1]
  2. The following data represents study hours (X) and scores (Y): X: 2, 4, 6, 8, 10; Y: 50, 55, 65, 70, 78. (i) Find the linear regression equation. (ii) Predict Y when X = 7. [CO2]
  3. Explain Bayesian modeling and the role of prior and posterior probabilities. [CO2]
  4. A data stream generates 120,000 records per minute. If only 35% data is retained using a decaying window, calculate the number of records retained per minute and justify the use of windowing. [CO3]
  5. Explain the importance of frequent itemset mining in market basket analysis. [CO4]

Section C

Q3 to Q7. Attempt any one part of each question · 7 × 5 = 35 marks

Q3

  1. Explain Support Vector Machines (SVM) with the concept of margin and support vectors. [CO2]
  2. A dataset has eigenvalues: 4.5, 2.8, 1.2, 0.4, 0.1. (i) Calculate percentage variance of each component. (ii) Determine the number of components required to retain 90% variance. [CO2]

Q4

  1. A sliding window of size 200 slides by 50 elements. Calculate the number of windows formed after processing 1,000 elements and explain its impact on latency. [CO3]
  2. Design a Real-Time Analytics Platform for social media sentiment analysis showing major components. [CO3]

Q5

  1. Given transactions: T1 {A, B, C}, T2 {A, B}, T3 {A, C}, T4 {B, C}, T5 {A, B, C}. If minimum support = 60%, find all frequent item sets and show pruning. [CO4]
  2. Compare hierarchical clustering and K-means clustering with suitable applications. [CO4]

Q6

  1. Explain the working of MapReduce with intermediate key–value pairs. [CO4]
  2. Design a Hadoop-based framework for storing and processing large-scale data. [CO4]

Q7

  1. For the dataset {12, 15, 20, 22, 25, 30}, calculate mean and standard deviation and comment on data spread. [CO5]
  2. Design an R-based workflow for analyzing unstructured text data. [CO5]

Most repeated topics

How many of the last 4 AKTU 5th-semester Data Analytics papers (KCS051 2022-23 and 2023-24, BCS052 2024-25 and 2025-26) asked about each topic.

Hadoop, MapReduce and Hive
4 / 4
Data streams (windows, sampling, filtering, counting)
4 / 4
Frequent itemsets, Apriori, market basket analysis
4 / 4
Clustering (K-means, hierarchical, CLIQUE/ProCLUS)
4 / 4
NoSQL vs relational databases
3 / 4
Bayesian methods and networks
3 / 4
Fuzzy logic
3 / 4
SVM and kernel methods
3 / 4
Data analytics lifecycle, types of data, Big Data
3 / 4
Regression
2 / 4
PCA and dimensionality reduction
2 / 4

Counted by LearnAIFast from the 2022-23, 2023-24, 2024-25 and 2025-26 odd-semester theory papers.

Answer hints

Q2(b). Linear regression for study hours

x̄ = 6, ȳ = 63.6. Σ(x − x̄)(y − ȳ) = 142 and Σ(x − x̄)² = 40, so slope b = 142 ÷ 40 = 3.55 and intercept a = 63.6 − 3.55 × 6 = 42.3. Equation: Y = 42.3 + 3.55X. At X = 7, Y = 42.3 + 24.85 = 67.15.

Q2(d). Decaying window, 35% retained

Records retained per minute = 0.35 × 120,000 = 42,000. Justify windowing: a stream is unbounded, so you can't store everything. A decaying window gives recent records more weight and lets old ones fade, which keeps memory fixed while still tracking current trends.

Q3(b). PCA variance from eigenvalues

Total = 4.5 + 2.8 + 1.2 + 0.4 + 0.1 = 9.0. Variance per component: 50%, 31.1%, 13.3%, 4.4%, 1.1%. Cumulative: 50%, 81.1%, 94.4%. So 3 components are needed to retain at least 90% variance.

Q4(a). Sliding window 200, slide 50, 1,000 elements

Number of full windows = ((1,000 − 200) ÷ 50) + 1 = 17. A smaller slide gives more frequent, fresher results but more computation per element. A larger window smooths results but adds latency before changes show up.

Q5(a). Frequent itemsets at 60% support

With 5 transactions, 60% support means a count of at least 3. Single items: A = 4, B = 4, C = 4, all frequent. Pairs: AB = 3, AC = 3, BC = 3, all frequent. Triple ABC = 2, not frequent, so it is pruned. Frequent itemsets: {A}, {B}, {C}, {A,B}, {A,C}, {B,C}.

Q7(a). Mean and standard deviation

Mean = 124 ÷ 6 ≈ 20.67. Squared deviations sum to about 215.33. Population SD = √(215.33 ÷ 6) ≈ 5.99; sample SD = √(215.33 ÷ 5) ≈ 6.56. State which one you use. The values are moderately spread around the mean with no outliers.

Other years

2024-25 (BCS052) · PDF 2023-24 (KCS051) · PDF 2022-23 (KCS051) · PDF 2021-22 (KCS051) · PDF 2020-21 (KCS051) · PDF All AI & ML papers

This paper is shared for study purposes only. Copyright belongs to AKTU. If you own the copyright and want it removed, email learnwithskillnest@gmail.com and we will take it down.