BCA 303T Artificial Intelligence Notes from Unnati Educations cover all 4 units of this IPU BCA Semester 5 paper in 75 pages: 40 solved two-mark questions, 33 ten-mark practice questions, 21 worked examples, 2 PROLOG programs with output and 14 figures. They follow the syllabus for the batch admitted in 2024-25 onwards.
Written by Prateek Talwar, Founder, Unnati Educations · Reviewed by Ritik Rawat, B.Tech, Senior Full Stack Developer · Checked against BCA Programme Common Scheme and Syllabus (w.e.f. A.S. 2024-25), ipu.ac.in, 3 Sep 2026 · Updated 13 October 2026 · Editorial policy
Scheme: BCA Programme Common Scheme and Syllabus (w.e.f. A.S. 2024-25) | Applies to: batch admitted 2024-25 onwards | Verified against: bcaaxV030926.pdf, 3 Sep 2026
On this page
BCA 303T Artificial Intelligence at a glance: GGSIPU BCA Sem 5 credits, units and exam pattern
BCA 303T is a core theory paper, listed with the other GGSIPU BCA Semester 5 papers; its only prerequisite is basic programming skills.
| Code and title | BCA 303T Artificial Intelligence |
|---|---|
| Course and semester | BCA, Semester 5 |
| Category and credits | Core Course Theory (CCT), 4 credits; L-T-C 4-0-4 (4 lecture hours a week) |
| Marks | 100: 40 internal + 60 end-term |
| Applies to | Batch admitted in 2024-25 onwards |
| End-term pattern | Q1: 10 compulsory 2-mark questions; then one of two 10-mark questions per unit (20 + 40 = 60 marks) |
| Units and teaching hours | 4 units: 12 + 12 + 10 + 10 = 44 hours |
| Paired lab | BCA 303P Artificial Intelligence Lab: CCP, 4 practical hours, 2 credits, 100 marks (40 + 60) |
Paper setters map every question to the four course outcomes: search (CO1), knowledge and inference (CO2), uncertainty (CO3) and systems that improve (CO4).
Grading follows Ordinance 11 on examinations and results.
BCA 303T Artificial Intelligence Notes unit-wise: 40 solved two-mark questions, 21 worked examples and 2 PROLOG programs across 4 units
Notes for the other papers sit under GGSIPU BCA notes.
BCA 303T Unit 1: Overview of AI, Problem Spaces and Heuristic Search
Unit 1 defines AI (four approaches, PEAS) and the Turing Test with Searle's Chinese Room. It frames problems as state space search (water jug, 8-puzzle, BFS vs DFS, production systems), then covers hill climbing, best-first, A*, AO* and constraint satisfaction problems.
Pages 5 to 23 · 3 sections · 5 worked examples, including SEND + MORE = MONEY · 10 solved two-mark, 8 ten-mark practice questions · 3 figures
| Feature | Greedy best-first | A* search |
|---|---|---|
| Evaluation function | f(n) = h(n) | f(n) = g(n) + h(n) |
| Optimal? | No | Yes, if h(n) is admissible |
| Complete? | No, it can loop | Yes, with finite branching and positive costs |
Quick look: greedy best-first search vs A* search (from page 18)
- Feature
- Evaluation function
- Greedy best-first
f(n) = h(n)- A* search
f(n) = g(n) + h(n)
- Feature
- Optimal?
- Greedy best-first
- No
- A* search
- Yes, if
h(n)is admissible
- Feature
- Complete?
- Greedy best-first
- No, it can loop
- A* search
- Yes, with finite branching and positive costs
BCA 303T Unit 2: Knowledge Representation and Logical Reasoning
Unit 2 covers the DIKW ladder, knowledge types and four representation approaches with semantic networks and frames. The logic half runs from propositional logic to first order logic, unification, forward and backward chaining on the Colonel West example and resolution in nine CNF steps, ending with a PROLOG expert system.
Pages 24 to 43 · 3 sections · 5 worked examples and 2 PROLOG programs with output · 10 solved two-mark, 8 ten-mark practice questions · 3 figures
Quick look: a grandparent rule in PROLOG (from page 38)
parent(dashrath, ram).
parent(kaushalya, ram).
parent(ram, lav).
grandparent(X, Z) :- parent(X, Y), parent(Y, Z).
?- grandparent(G, lav).
% G = dashrath ; G = kaushalya.Backward chaining proves the rule; backtracking finds the second answer.
BCA 303T Unit 3: Handling Uncertainty and Natural Language Processing
Unit 3 covers non-monotonic reasoning (default logic, closed world assumption, truth maintenance), Bayes' theorem, certainty factors (see the excerpt below), the burglary-alarm Bayesian network and Dempster-Shafer belief intervals. Fuzzy logic ends with an AC controller design; NLP covers five stages, ambiguity and parsing.
Pages 44 to 60 · 2 sections · 7 worked examples · 10 solved two-mark, 8 ten-mark practice questions · 4 figures
Quick look: Bayes' theorem on a medical test (from page 47)
- Given:
P(D) = 0.01,P(+|D) = 0.95,P(+|¬D) = 0.05. - Total evidence:
P(+) = 0.95 × 0.01 + 0.05 × 0.99 = 0.059. - Posterior:
P(D|+) = 0.0095 / 0.059 = 0.161, about 16%.
BCA 303T Unit 4: Learning
Unit 4 covers rote learning (Samuel's checkers), taking advice (Mostow's FOO), parameter adjustment, macro-operators and chunking, induction through Winston's ARCH, version spaces and ID3, explanation-based learning, discovery, analogy, neural networks and genetic algorithms.
Pages 61 to 75 · 1 section with 9 sub-sections · 4 worked examples, including ID3 on Play Tennis · 10 solved two-mark, 9 ten-mark practice questions · 4 figures
| Measure | Initial population | After one generation |
|---|---|---|
| Sum of fitness | 1170 | 1754 |
| Average fitness | 292.5 | 438.5 |
| Best chromosome (x, fitness) | 11000 (24, 576) | 11011 (27, 729) |
Quick look: one generation of a genetic algorithm for f(x) = x² (from page 72)
- Measure
- Sum of fitness
- Initial population
- 1170
- After one generation
- 1754
- Measure
- Average fitness
- Initial population
- 292.5
- After one generation
- 438.5
- Measure
- Best chromosome (x, fitness)
- Initial population
- 11000 (24, 576)
- After one generation
- 11011 (27, 729)
Certainty factors and rule-based systems in AI: excerpt from page 48 of the BCA 303T notes
Excerpt from page 48
1.4 Certainty factors and rule-based systems
Certainty factors (CF) were introduced in MYCIN (Shortliffe and Buchanan, Stanford) as a practical alternative to full probability theory. Experts find it easier to say "this rule makes me fairly sure" than to give exact probabilities.
| Measure | Meaning | Range |
|---|---|---|
MB(H, E) | Measure of belief: increase in belief in H due to evidence E | 0 to 1 |
MD(H, E) | Measure of disbelief: increase in disbelief in H due to E | 0 to 1 |
CF(H, E) | Certainty factor = MB(H, E) − MD(H, E) | −1 (definitely false) to +1 (definitely true) |
Measures used with certainty factors
- Measure
MB(H, E)- Meaning
- Measure of belief: increase in belief in H due to evidence E
- Range
- 0 to 1
- Measure
MD(H, E)- Meaning
- Measure of disbelief: increase in disbelief in H due to E
- Range
- 0 to 1
- Measure
CF(H, E)- Meaning
- Certainty factor =
MB(H, E) − MD(H, E) - Range
- −1 (definitely false) to +1 (definitely true)
A rule carries its own CF. For example: IF the organism is gram-positive AND grows in chains THEN it is streptococcus, CF 0.7.
Combining certainty factors
| Situation | Formula |
|---|---|
| Premise with AND | CF(P1 AND P2) = min(CF(P1), CF(P2)) |
| Premise with OR | CF(P1 OR P2) = max(CF(P1), CF(P2)) |
| Rule propagation | CF(conclusion) = CF(rule) × CF(premise) |
| Two rules, both CF > 0 | CF = CF1 + CF2 × (1 − CF1) |
| Two rules, both CF < 0 | CF = CF1 + CF2 × (1 + CF1) |
| Two rules, opposite signs | CF = (CF1 + CF2) / (1 − min(|CF1|, |CF2|)) |
Rules for combining certainty factors
- Situation
- Premise with AND
- Formula
CF(P1 AND P2) = min(CF(P1), CF(P2))
- Situation
- Premise with OR
- Formula
CF(P1 OR P2) = max(CF(P1), CF(P2))
- Situation
- Rule propagation
- Formula
CF(conclusion) = CF(rule) × CF(premise)
- Situation
- Two rules, both CF > 0
- Formula
CF = CF1 + CF2 × (1 − CF1)
- Situation
- Two rules, both CF < 0
- Formula
CF = CF1 + CF2 × (1 + CF1)
- Situation
- Two rules, opposite signs
- Formula
CF = (CF1 + CF2) / (1 − min(|CF1|, |CF2|))
Solved examples
| Question | Working | Answer |
|---|---|---|
| Rule CF = 0.8; premise "A AND B" with CF(A) = 0.9, CF(B) = 0.7 | 0.8 × min(0.9, 0.7) = 0.8 × 0.7 | 0.56 |
| Two rules support H with CF1 = 0.6, CF2 = 0.4 | 0.6 + 0.4 × (1 − 0.6) = 0.6 + 0.16 | 0.76 |
| A third rule adds CF = 0.5 to the result 0.76 | 0.76 + 0.5 × (1 − 0.76) = 0.76 + 0.12 | 0.88 |
| CF1 = 0.6 (for), CF2 = −0.4 (against) | (0.6 − 0.4) / (1 − 0.4) = 0.2 / 0.6 | 0.33 |
| CF1 = −0.5, CF2 = −0.3 | −0.5 + (−0.3) × (1 − 0.5) = −0.5 − 0.15 | −0.65 |
Certainty factor calculations worked on page 48
- Question
- Rule CF = 0.8; premise "A AND B" with CF(A) = 0.9, CF(B) = 0.7
- Working
- 0.8 × min(0.9, 0.7) = 0.8 × 0.7
- Answer
- 0.56
- Question
- Two rules support H with CF1 = 0.6, CF2 = 0.4
- Working
- 0.6 + 0.4 × (1 − 0.6) = 0.6 + 0.16
- Answer
- 0.76
- Question
- A third rule adds CF = 0.5 to the result 0.76
- Working
- 0.76 + 0.5 × (1 − 0.76) = 0.76 + 0.12
- Answer
- 0.88
- Question
- CF1 = 0.6 (for), CF2 = −0.4 (against)
- Working
- (0.6 − 0.4) / (1 − 0.4) = 0.2 / 0.6
- Answer
- 0.33
- Question
- CF1 = −0.5, CF2 = −0.3
- Working
- −0.5 + (−0.3) × (1 − 0.5) = −0.5 − 0.15
- Answer
- −0.65
Key point: combining two positive CFs always gives a bigger CF, but it can never exceed 1. More supporting evidence increases belief, with diminishing returns.
Limitations: CFs have no strict probabilistic meaning, results can depend on the order of combination in complex rule chains, and they assume rules are independent. Even so, MYCIN performed as well as human experts in tests.
BCA 303T exam corner: 40 solved two-mark and 33 ten-mark practice questions by unit
| Unit | 2 Marks (solved) | 10 Marks (practice) |
|---|---|---|
| Unit 1 | 10 | 8 |
| Unit 2 | 10 | 8 |
| Unit 3 | 10 | 8 |
| Unit 4 | 10 | 9 |
| Total | 40 | 33 |
BCA 303T exam questions in the notes, by unit
- Unit
- Unit 1
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 8
- Unit
- Unit 2
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 8
- Unit
- Unit 3
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 8
- Unit
- Unit 4
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 9
- Unit
- Total
- 2 Marks (solved)
- 40
- 10 Marks (practice)
- 33
Two-mark questions have model answers; ten-mark ones are practice prompts. One question per unit:
- Unit 1: What is a plateau in hill climbing? A flat region where all neighbours score the same; remedy, a big jump.
- Unit 2: What is Skolemization? CNF step 5: replace each ∃ variable with a Skolem constant or function.
- Unit 3: Define belief and plausibility in Dempster-Shafer theory.
Bel(S)sums the mass of subsets of S;Pl(S) = 1 − Bel(¬S). - Unit 4 (ten-mark): Explain the perceptron and its learning rule, and train a perceptron for the AND gate. Use
wi = wi + η(d − y)xi; the worked weights settle at (0.1, 0.1).
The BCA 303T notes solve 40 two-mark questions across 4 units and set 33 ten-mark practice questions, plus 21 worked examples and 2 PROLOG programs with output.
More practice for this paper sits in the BCA 303T important questions.
Who writes and checks the BCA 303T AI notes?
- Prateek Talwar, author of the BCA 303T notes · Founder, Unnati Educations · Associate Company Secretary (ACS), ICSI, since 15 May 2019 · B.Com, University of Delhi (School of Open Learning), 2012 · Profile
- Ritik Rawat, subject reviewer for the BCA 303T notes · Senior Full Stack Developer, Unnati Educations · B.Tech · Profile
The 2026-27 edition of the BCA 303T Artificial Intelligence IPU notes is mapped to the BCA scheme for the batch admitted in 2024-25 onwards, whose third-year syllabus was approved in August 2026.
It has nine topic sections, four exam corners and four quick revision tables; updated 13 October 2026 (editorial policy).
The same team prepares GGSIPU important questions.
Do the BCA 303T notes cover the full GGSIPU syllabus?
Yes. Every topic in the four units of the official BCA 303T syllabus has its own section; a topic-by-topic check found no gaps.
| Official unit | Official topics | Where in the notes | Status |
|---|---|---|---|
| Unit I (12 hours) | Overview of AI; problems, problem space and search (state space, production systems, search design issues); heuristic search (generate and test, hill climbing, best first, problem reduction, constraint satisfaction) | Sections 1 to 3, p6 to 20 | Covered |
| Unit II (12 hours) | Knowledge representation (definition, approaches, issues); logical reasoning (logical agents, propositional logic, FOL syntax, semantics and inference, forward and backward chaining, unification, resolution); expert system case study in PROLOG | Sections 1 to 3, p25 to 40 | Covered |
| Unit III (10 hours) | Non-monotonic and probabilistic reasoning, Bayes' theorem, certainty factors, Bayesian networks, Dempster-Shafer theory, fuzzy logic, fuzzy sets and types, membership functions, fuzzy set design; NLP (syntactic, semantic, pragmatic processing) | Sections 1 to 2, p45 to 57 | Covered |
| Unit IV (10 hours) | Rote learning, taking advice, learning in problem solving, induction, explanation-based learning, discovery, analogy, neural networks, genetic learning | Section 1 (1.1 to 1.9), p62 to 72 | Covered |
Official BCA 303T syllabus mapped to the notes
- Official unit
- Unit I (12 hours)
- Official topics
- Overview of AI; problems, problem space and search (state space, production systems, search design issues); heuristic search (generate and test, hill climbing, best first, problem reduction, constraint satisfaction)
- Where in the notes
- Sections 1 to 3, p6 to 20
- Status
- Covered
- Official unit
- Unit II (12 hours)
- Official topics
- Knowledge representation (definition, approaches, issues); logical reasoning (logical agents, propositional logic, FOL syntax, semantics and inference, forward and backward chaining, unification, resolution); expert system case study in PROLOG
- Where in the notes
- Sections 1 to 3, p25 to 40
- Status
- Covered
- Official unit
- Unit III (10 hours)
- Official topics
- Non-monotonic and probabilistic reasoning, Bayes' theorem, certainty factors, Bayesian networks, Dempster-Shafer theory, fuzzy logic, fuzzy sets and types, membership functions, fuzzy set design; NLP (syntactic, semantic, pragmatic processing)
- Where in the notes
- Sections 1 to 2, p45 to 57
- Status
- Covered
- Official unit
- Unit IV (10 hours)
- Official topics
- Rote learning, taking advice, learning in problem solving, induction, explanation-based learning, discovery, analogy, neural networks, genetic learning
- Where in the notes
- Section 1 (1.1 to 1.9), p62 to 72
- Status
- Covered
Official source: BCA Programme Common Scheme and Syllabus, batch admitted 2024-25 onwards (ipu.ac.in).
Every course listed under GGSIPU notes is mapped to its official PDF in the same way.
How to write BCA 303T answers: two-mark definitions, ten-mark numericals and logic proofs
Q1 takes at least two answers from every unit. The solved two-mark answers show the length: a definition plus one example or formula.
For ten-mark answers, the notes' exam tips list what to include:
- Turing Test: definition, three participants, abilities needed and the Chinese Room criticism (page 11).
- Fuzzy logic: state how it differs from probability (page 51).
- Unification: show the occurs check, where x cannot unify with f(x) (page 33).
The notes flag the commonest logic slip: ∀ goes with →, and ∃ goes with ∧ (page 31). Worked example:
- "Not all students like Maths": write the ∀ and → form, then negate it:
¬∀x (student(x) → like(x, Maths)). - Equivalent form with ∃ and ∧:
∃x (student(x) ∧ ¬like(x, Maths)), some student does not like Maths.
For numericals, write the formula, substitute, then box the answer. Practise this format on the GGSIPU BCA important questions.
BCA 303T and the BCA 303P AI lab, plus other GGSIPU BCA Sem 5 papers
BCA 303P AI lab notes: the practical component of this paper (CCP, 4 practical hours a week, 2 credits, 100 marks). Its practical list is set by the teacher in week one.
Unit 2's two PROLOG programs are the closest bridge to the lab; Unit 1 search methods and the Unit 4 perceptron and genetic algorithm are step-by-step algorithms, easy to code.
Other Semester 5 papers: BCA 301T Computer Networks and BCA 301P lab; one elective from BCA 305T Natural Language Processing, BCA 307T Network Security or BCA 309T Full Stack Development using Java, with its lab; BCA 311 Summer Training; and BCA 313T Introduction to Management and Entrepreneurship Development or BCA 315T MOOC Course.
If you pick BCA 305T, the Unit 3 NLP pages here are a useful first pass.
BCA 303T AI notes FAQs
Which BCA 303T unit has the most worked examples?
Unit 3 has the most, with 7 worked examples: two Bayes' theorem problems, a five-row certainty factor table, the burglary alarm network, a Dempster-Shafer combination, fuzzy set operations and the fuzzy AC controller. Unit 1 and Unit 2 have 5 each, and Unit 2 adds both PROLOG programs. Unit 4 has 4. Unit 2 is the longest unit at 20 pages.
Do the BCA 303T notes include PROLOG code?
Yes. Unit 2 has two PROLOG programs with SWI-Prolog output. The first is a family knowledge base with father, mother, grandparent, sibling and a recursive ancestor rule. The second is an animal identification expert system, the PROLOG case study the BCA 303T syllabus asks for, with a table linking each program part to expert system architecture.
Do these BCA 303T notes match my batch?
They match the BCA 303T syllabus for students admitted in 2024-25 onwards, as printed in the BCA Programme Common Scheme and Syllabus on ipu.ac.in. If your batch follows an earlier BCA scheme, compare its Artificial Intelligence syllabus with the four units listed here before you order; your college can confirm which scheme applies to you.
What is the passing mark for BCA 303T?
BCA 303T is marked out of 100: 40 internal and 60 end-term. The BCA scheme PDF does not print a pass percentage for theory papers; its 40% pass line applies only to a first-semester pass/fail paper. End-term minimums, grading and re-appear rules are set by Ordinance 11 on ipu.ac.in. Your college can confirm the rule for your batch.
Get the full BCA 303T Artificial Intelligence Notes on WhatsApp
The full BCA 303T Artificial Intelligence Notes are a printed book of all 75 pages, not a PDF.
| Part | Pages | 2 Marks (solved) | 10 Marks (practice) |
|---|---|---|---|
| Cover, overview and index | 4 (p1 to 4) | - | - |
| Unit 1 | 19 (p5 to 23) | 10 | 8 |
| Unit 2 | 20 (p24 to 43) | 10 | 8 |
| Unit 3 | 17 (p44 to 60) | 10 | 8 |
| Unit 4 | 15 (p61 to 75) | 10 | 9 |
| Total | 75 | 40 | 33 |
Pages and questions per unit in the BCA 303T book
- Part
- Cover, overview and index
- Pages
- 4 (p1 to 4)
- 2 Marks (solved)
- -
- 10 Marks (practice)
- -
- Part
- Unit 1
- Pages
- 19 (p5 to 23)
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 8
- Part
- Unit 2
- Pages
- 20 (p24 to 43)
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 8
- Part
- Unit 3
- Pages
- 17 (p44 to 60)
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 8
- Part
- Unit 4
- Pages
- 15 (p61 to 75)
- 2 Marks (solved)
- 10
- 10 Marks (practice)
- 9
- Part
- Total
- Pages
- 75
- 2 Marks (solved)
- 40
- 10 Marks (practice)
- 33
Unnati Educations is an independent education provider based in Vikaspuri, Delhi. We are not affiliated with, authorised by, or endorsed by Guru Gobind Singh Indraprastha University (GGSIPU). 'GGSIPU' and 'IPU' are used only to describe the syllabus our notes follow. For official information, visit ipu.ac.in.
Ready for the full notes?
Discuss coverage and delivery with our team.