GTU · BE IT · Semester 7
GTU 3171613 Pattern Recognition Old Question Papers PDF
Official Gujarat Technological University (GTU) old question papers PDF (Regular & Remedial) and step-by-step solved questions for Pattern Recognition across 2025 – 2026 examinations.
18 questions GTU asked more than onceAsked in all 3 papers: Explain Decision Tree learning based on the Classification and Regression Trees (CART) approach with example.See listExam papers
Most repeated questions
18 questions GTU set in two or more of 3 papers, most repeated first. Tap a year to open that paper at the question.
Explain Decision Tree learning based on the Classification and Regression Trees (CART) approach with example.
State the main idea behind Support Vector Machines (SVM) for binary classification.
Explain the steps in “k-means clustering” using a suitable illustration. List out limitations of “k-means clustering”.
Explain Hidden Markov model (HMMs) and its role in the classifier design.
Explain the Expectation-Maximization (EM) algorithm in the context of Gaussian Mixture Models (GMMs) and demonstrate how it can be used to estimate the parameters of a GMM.
Discuss multilayer feed-forward neural networks with neat architecture diagram
Discuss the principles and applications of nonparametric techniques for density estimation, focusing on the K-Nearest Neighbors (K-NN) method. Provide examples of how K-NN can be employed in practice.
Difference between joint probability and conditional probability
Explain Fisher Discriminant Analysis (FDA) in the context of supervised dimensionality reduction.
Explain with an example why large dimension of dataset is not desirable ?
Show 8 more repeated questionsHide the 8 more
Define Gradient Descent, and how is it used in optimizing linear discriminant functions.
Explain working of Recurrent Neural Networks (RNN).
Describe the key differences between Bayesian estimate and maximum likelihood estimation.
Consider two class classification problem. The classes are denoted as w1 and w2. Apply Baye’s minimum risk classification rule to decide whether sample x belongs to class w1 or w2. Use the following data: P(w1) = 0.6, P(w2) = 0.4, P(x) = 0.74, P(x|w1) = 0.75, P(x|w2) = 0.4. Loss parameters are: λ11 = 0.3, λ12 = 0.7, λ21 = 0.65, λ22 = 0.35.
State Bayes Theorem and explain its significance.
Explain eigenvalues and eigenvectors in brief.
Define nominal data in the context of pattern classification.
Discuss idea of pooling operating used in Convolutional Neural Networks (CNNs) with an example.
Which branches take this subject
GTU's syllabus lists 3171613 for 2 branches. One set of papers serves all of them.
Semester 7, elective: 2 branches
Computer Science & Engineering (Data Science), Information Technology
Other Semester 7 subjects
Common questions
Are these GTU 3171613 Pattern Recognition papers valid for both Regular and Remedial exams?
Yes. Gujarat Technological University (GTU) follows the same 70-mark syllabus, exam blueprint, and 23-mark passing cutoff for both Regular and Remedial (ATKT/backlog) examinations. All 3 past exam sessions (2025 – 2026) serve as comprehensive preparation for either attempt.
How can I download GTU 3171613 Pattern Recognition old question papers PDF?
You can download official GTU 3171613 original question paper PDFs directly from the table above with 1 click and zero deceptive ads or wait timers. Every paper also includes interactive online questions with AI study prompts.
What is the passing mark for GTU Pattern Recognition (3171613)?
The external theory exam carries 70 marks. The minimum passing cutoff is 23 marks (33%). Exam duration is 2.5 hours.
What is the GTU exam pattern for 3171613?
The examination contains 5 main questions of 14 marks each, totaling 70 marks. Questions 2 through 5 provide internal OR choices, typically split into 3-mark, 4-mark, and 7-mark questions.