The GATE DA 2027 syllabus is the single most important document for any Data Science & AI aspirant preparing for GATE 2027. Every hour of your preparation should be anchored to it. This guide provides the complete, official GATE DA 2027 syllabus across all 7 core subjects — with exact topics, expected weightage, preparation tips, and a clear breakdown of what matters most for a high rank. Whether you are starting your preparation from scratch or doing a final syllabus audit before the exam, this is the most comprehensive reference available.
GATE DA 2027 — Exam Overview & Paper Pattern
GATE DA (Data Science & Artificial Intelligence) 2027 is a 3-hour, 100-mark computer-based test with 65 questions. The paper is expected to be organised by IIT Madras for GATE 2027. Understanding the paper structure is essential before diving into the syllabus — it tells you how many marks each section is fighting for.
GATE DA 2027 Syllabus — Subject-Wise Weightage at a Glance
The table below shows the expected weightage for each subject in the GATE DA 2027 syllabus, derived from analysis of GATE DA papers from 2021 to 2026. Use this as your strategic allocation guide for preparation time.
| # | Subject | Expected Marks | Weightage | Priority |
|---|---|---|---|---|
| — | General Aptitude | 15 (fixed) | Fixed | |
| 1 | Probability & Statistics | ~17–20 | Very High | |
| 2 | Linear Algebra | ~11–14 | High | |
| 3 | Calculus & Optimization | ~9–11 | High | |
| 4 | Programming, DSA | ~8–11 | High | |
| 5 | DBMS & Data Warehousing | ~7–10 | Medium | |
| 6 | Machine Learning | ~18–22 | Very High | |
| 7 | Artificial Intelligence | ~9–12 | High |
Subject 1 — Probability & Statistics
The mathematical spine of all Data Science. Every ML algorithm has a probabilistic foundation — and GATE DA tests that foundation directly. This is the subject where most aspirants either win or lose their rank. The official GATE DA 2027 syllabus lists the following topics:
🎯 Preparation Strategy — Probability & Statistics
- Master the big 5 distributions (Normal, Binomial, Poisson, t, Chi-squared) — their PMF/PDF, mean, variance, and when each applies. Questions test both formula recall and applied judgement.
- Bayes Theorem appears every year — often in disguise as an ML question. Practice deriving posterior probabilities from prior and likelihood.
- CLT is tested with numerical problems: given a distribution and sample size, compute probabilities on sample means using Normal approximation.
- Hypothesis testing questions require you to identify the correct test (z vs t vs chi-squared), compute the test statistic, and state the conclusion. Practice all three types.
- Note: MLE and MAP are not explicitly listed in the Probability section but appear in the Machine Learning section — master them as core connecting concepts.
Subject 2 — Linear Algebra
Linear Algebra is both a standalone GATE DA section and the hidden language underneath Machine Learning. PCA, SVM, neural networks, and regression — all lean on linear algebraic foundations. The official GATE DA 2027 syllabus for Linear Algebra covers:
🎯 Preparation Strategy — Linear Algebra
- SVD is the most important topic. Understand it geometrically (what it decomposes), computationally (how to compute from eigenvalues), and in ML context (its connection to PCA and low-rank approximations).
- Eigenvalues/eigenvectors: Practice computation for 2×2 and 3×3 matrices, and understand their properties (trace = sum of eigenvalues, det = product of eigenvalues, symmetric matrices have real eigenvalues).
- Rank and nullity: Use rank-nullity theorem, understand full rank vs rank-deficient matrices, and how rank relates to solution existence in linear systems.
- Projection matrices appear in the context of least squares regression — understand P = X(X'X)⁻¹X' and its idempotent, symmetric properties.
- Quadratic forms (x'Ax) — know positive definite/semidefinite conditions and their role in optimization and covariance matrices.
Subject 3 — Calculus & Optimization
The GATE DA 2027 syllabus keeps this section focused — single-variable calculus only, with an emphasis on optimization. This makes it one of the more tractable sections to master completely. Official topics:
🎯 Preparation Strategy — Calculus & Optimization
- Memorise Taylor/Maclaurin series for e^x, sin x, cos x, ln(1+x), (1+x)^n — these appear in ML-connected questions and directly in optimization.
- Practice identifying maxima, minima, and saddle points using first and second derivative tests, including cases where the test is inconclusive.
- This section has the highest marks-per-study-hour ratio for most aspirants — the scope is tight and the topics are predictable. Target 100% mastery here.
Subject 4 — Programming, Data Structures & Algorithms
Unlike GATE CS where DSA is tested at great depth, GATE DA 2027 tests DSA and Programming at a conceptual and applied level — enough for a data scientist to write and analyse code. The official GATE DA 2027 syllabus covers:
🎯 Preparation Strategy — Programming & DSA
- Python is explicitly named in the syllabus. Questions test Python code tracing, output prediction, and function logic — not deep software engineering. Practice reading and tracing short Python code snippets.
- Time complexity analysis is the most common question type here. Know the complexity of all listed sorts (O(n log n) for merge/quick on average), BFS/DFS (O(V+E)), Dijkstra (O((V+E) log V)), and binary search (O(log n)).
- Graph traversal questions often overlap with the AI section — BFS/DFS appear in both contexts (DSA for data structures context, AI for search algorithms context).
- Dijkstra's algorithm is tested both algorithmically (trace through a given graph) and conceptually (when it fails — negative weights).
Subject 5 — Database Management & Warehousing
One of the most practical sections for a working data scientist — and one of the easiest to score full marks in with structured preparation. The official GATE DA 2027 syllabus for DBMS covers:
🎯 Preparation Strategy — DBMS & Data Warehousing
- SQL questions require you to write/trace queries with JOINs (INNER, LEFT, RIGHT), GROUP BY, HAVING, subqueries, and aggregate functions. Practice translating English descriptions to SQL.
- Normalisation (1NF → 2NF → 3NF → BCNF) questions give you a relation with functional dependencies and ask you to identify the normal form or decompose it. This is a reliable 2-mark question source every year.
- B+ Trees: Understand insertion, deletion, and the difference from B-Trees (all data in leaves for B+). Questions often test tree structure after insertions.
- Star vs Snowflake schema — straightforward conceptual distinction but reliably tested in Data Warehousing. Know the trade-offs (query performance vs storage).
Master the Complete GATE DA 2027 Syllabus with IIT Madras Alumni Faculty
The PiyushAI GATE DA 2027 Course covers every topic in this syllabus with dedicated lectures, PYQ analysis, and GATE-level mock tests — at ₹6,500.
🚀 Enroll in GATE DA 2027 Course — ₹6,500Subject 6 — Machine Learning
The highest-weightage subject in GATE DA — and the most numerically demanding. Machine Learning in GATE DA 2027 is split into Supervised and Unsupervised Learning. This is where IIT-level conceptual depth most visibly separates top rankers. The official GATE DA 2027 syllabus for ML:
Supervised Learning
Unsupervised Learning
🎯 Preparation Strategy — Machine Learning
- Logistic Regression: Know the sigmoid, log-loss (cross-entropy), gradient derivation, and how regularization (L2/Ridge) affects the decision boundary.
- SVM: Understand margin = 2/‖w‖, the role of support vectors, and the effect of the C parameter. Hard margin vs soft margin. Linear vs kernel SVM.
- Decision Trees: Compute information gain (entropy-based) and Gini impurity. Know when to stop splitting (pure node, max depth). Trace a small tree by hand.
- PCA: Know that PCA = eigendecomposition of covariance matrix = truncated SVD of data matrix. Be able to determine the number of components from explained variance.
- Bias-Variance: Understand the decomposition mathematically — bias², variance, and irreducible error. Know how regularization, tree depth, and k in k-NN affect each.
- Neural Networks: Forward pass computation, backpropagation chain rule, activation functions (sigmoid, ReLU, tanh) and their derivatives. Questions often give a small network and ask for output or gradient.
Subject 7 — Artificial Intelligence
The AI section of GATE DA 2027 covers classical AI — search, logic, and probabilistic reasoning. It is conceptually rich and uniquely tested compared to GATE CS. The official GATE DA 2027 syllabus for AI:
🎯 Preparation Strategy — Artificial Intelligence
- A* Search: Understand heuristic admissibility (h(n) ≤ true cost) and consistency (monotonicity). Practice tracing A* on small graphs and identifying whether a given heuristic is admissible.
- Minimax + Alpha-Beta: Trace minimax on game trees by hand. Alpha-Beta pruning: track α and β values as the tree expands, and identify which branches get pruned. A very reliable GATE DA question type.
- Bayesian Networks: Draw conditional independence graphs, compute joint probabilities using the chain rule, and run variable elimination by marginalisation. These questions reward careful probability arithmetic.
- Logic: Propositional calculus — truth tables, CNF/DNF, resolution. First-order predicate logic — universal/existential quantifiers, unification. Practice converting English sentences to first-order formulas.
- Approximate inference (MCMC sampling) — understand the concept of sampling-based approximation; typically tested at a high level rather than with detailed algorithm tracing.
General Aptitude — Fixed 15 Marks
General Aptitude is common to all GATE papers and carries a fixed 15 marks (10 questions: 5 × 1 mark + 5 × 2 marks). It has two sub-sections:
- Verbal Ability: English grammar, sentence correction, vocabulary, reading comprehension, verbal analogies.
- Numerical Ability: Arithmetic (percentages, ratios, averages, profit/loss), data interpretation (tables, charts), number series, basic algebra, and logical reasoning.
Many GATE DA aspirants neglect General Aptitude, treating it as "easy" and under-preparing. In reality, a full 15/15 in General Aptitude — which is achievable with 2–3 weeks of focussed practice — can be the difference between AIR 50 and AIR 200. Never leave this section to chance.
What Is NOT in the Official GATE DA 2027 Syllabus
Just as important as knowing what to study is knowing what to not waste time on. Here is a clear list of topics that aspirants frequently study but are not in the official GATE DA 2027 syllabus:
- Multivariate calculus / partial derivatives (as a standalone section)
- Complex numbers & complex analysis
- Differential equations
- Numerical methods (Newton-Raphson etc.)
- CNNs, RNNs, LSTMs, Transformers (deep architectures beyond MLP)
- Reinforcement Learning
- Natural Language Processing (NLP)
- Computer Vision beyond basic CNN conceptuals
- Operating Systems, Computer Networks, Computer Architecture
- Compiler Design, Theory of Computation
- Advanced SQL (stored procedures, triggers, transactions — barely tested)
- Gradient descent (via ML section — not Calculus section)
- MLP / feed-forward neural networks
- Bayesian networks & variable elimination (AI section)
- Dijkstra's algorithm (DSA section)
- B+ Trees (DBMS section)
- Chi-squared test (Probability section)
- Ridge Regression / L2 regularisation (ML section)
- PCA via SVD (ML section — links to Linear Algebra)
- Alpha-Beta pruning (AI section)
- Divide & conquer sorting (DSA section)
How to Use the GATE DA 2027 Syllabus for Preparation
The GATE DA 2027 syllabus is not just a topic checklist — it is a strategic document. Here is how to use it correctly:
- Use it as your progress tracker: Print this syllabus. After completing every topic in your GATE DA 2027 course, physically tick it. This creates closure and prevents "I thought I covered this" surprises before the exam.
- Weight your time allocation: Based on the weightage table — allocate time proportionally. ML + Statistics alone deserve ~40% of your total preparation time. Do not treat all subjects equally.
- Pair every topic with PYQs: After covering any topic, immediately solve all GATE DA 2021–2026 questions on that topic. The PYQ is the ground truth for what "tested" actually means.
- Use "hot topics" as exam anchors: The 🔥-marked topics in each subject above are the ones that appear with highest frequency and highest marks in GATE DA. If time-constrained, ensure these are bulletproof before expanding to the rest.
- Audit the "Not in Syllabus" list regularly: Many coaching resources include out-of-scope topics. If you encounter a topic that is not in the official syllabus, note it and deprioritise it ruthlessly.
Frequently Asked Questions — GATE DA 2027 Syllabus
The official GATE DA 2027 syllabus covers 7 core subjects: (1) Probability & Statistics, (2) Linear Algebra, (3) Calculus & Optimization, (4) Programming, Data Structures & Algorithms, (5) Database Management & Warehousing, (6) Machine Learning (Supervised + Unsupervised), and (7) Artificial Intelligence (Search, Logic, Probabilistic Reasoning) — plus General Aptitude carrying a fixed 15 marks.
Machine Learning and Probability & Statistics have the highest weightage in GATE DA 2027, together accounting for approximately 38–42% of the total marks. Machine Learning alone carries ~21% and Probability & Statistics carries ~19%. These two subjects are the core of any serious GATE DA preparation strategy.
Yes. Linear Algebra is highly important for GATE DA 2027, both as a standalone section (~13% weightage) and as the mathematical foundation for ML topics like PCA (SVD), SVM (dot products in feature space), and neural networks (matrix operations). The syllabus covers vector spaces, eigenvalues, SVD, LU decomposition, and projections.
The official GATE DA 2027 syllabus includes multi-layer perceptron (MLP) and feed-forward neural networks under Machine Learning. CNNs, RNNs, LSTMs, and Transformers are NOT listed. Neural network fundamentals — forward pass, backpropagation, activation functions — are in scope. Focus on MLP depth rather than breadth across deep architectures.
The GATE DA 2027 syllabus explicitly lists Python as the programming language for the Programming, Data Structures & Algorithms section. Candidates should be comfortable reading, tracing, and understanding Python code, as well as implementing standard data structures and algorithms conceptually in Python.
Yes. For MCQ questions in GATE DA 2027: 1/3 mark negative for 1-mark questions and 2/3 mark negative for 2-mark questions. NAT (Numerical Answer Type) questions carry no negative marking. Always attempt all NAT questions. For MCQs, apply a ~80% confidence threshold before marking.
The PiyushAI Complete GATE DA 2027 Course & Test Series covers 100% of the official GATE DA syllabus across all 7 subjects with dedicated courses for each, 250+ hours of video, 4,000+ practice questions, full-length mock tests, AI analytics, and the Data Science Foundation Certificate — at ₹6,500. Taught by Piyush Wairale, an IIT Madras Alumnus. GATE DA AIR 2 rankers have credited this course. Enroll at www.piyushwairale.com.
Now You Know the Syllabus — It's Time to Master It
All 7 subjects. 250+ hours. 4,000+ questions. Full mock tests. AI analytics. Data Science Foundation Certificate. One IIT Madras Alumni instructor.
✅ Enroll & Get Certified — ₹6,500 →
Piyush Wairale is an IIT Madras Alumnus and former instructor at the IIT Madras BS Data Science Degree Programme. Founder of PiyushAI Edtech, Amravati, Maharashtra. Mentored 10,000+ students including GATE DA AIR 2 rankers. Microsoft Learn Educator · NPTEL+ Speaker · NVIDIA AI Summit Speaker · Ex-AWS Academy · Ex-Volkswagen.