ISC YEAR 2027

INDIAN SCHOOL CERTIFICATE

EXAMINATION

ARTIFICIAL
INTELLIGENCE
(883)
February 2025
____________________________________________________________________________________________

© Copyright, Council for the Indian School Certificate Examinations

All rights reserved. The copyright to this publication and any part thereof solely vests in the Council for the Indian
School Certificate Examinations. This publication and no part thereof may be reproduced, transmitted, distributed or
stored in any manner whatsoever, without the prior written approval of the Council for the Indian School Certificate
Examinations.
Council for the Indian School Certificate Examinations (CISCE)

MISSION STATEMENT

The Council for the Indian School Certificate

Examinations is committed to serving the nation's
children, through high quality educational
endeavours, empowering them to contribute towards
a humane, just and pluralistic society, promoting
introspective living, by creating exciting learning
opportunities, with a commitment to excellence.

ETHOS OF CISCE

Trust and fair play.

Minimum monitoring.
Allowing schools to evolve their own niche.
Catering to the needs of the children.
Giving freedom to experiment with new ideas
and practices.
Diversity and plurality - the basic strength for
evolution of ideas.
Schools to motivate pupils towards the
cultivation of:
Excellence - The Indian and Global
experience.
Values - Spiritual and cultural - to be the bedrock
of the educational experience.
Schools to have an 'Indian Ethos', strong roots in
the national psyche and be sensitive to national
aspirations.
 ARTIFICIAL INTELLIGENCE (883)

This subject may be taken with Computer Science but not with Robotics.

Aims
1. To develop an understanding of concepts and through learning and engaging in hands-on
applications in AI. activities.
2. To develop competencies in AI via classroom 4. To introduce concepts of data modelling.
instruction, laboratory and self-directed project- 5. To facilitate appreciation, understanding and
based learning approach. application of concepts of neural networks and
3. To facilitate appreciation, understanding and natural language processing.
application of concepts of data science in AI 6. To introduce concepts of generating predictions
from data.

CLASS XI
There will be two papers in the subject:
Paper II: Practical Exam - 3 hours ... 15 marks
Paper I: Theory - 3 hours ... 70 marks
Project Work … 10 marks
Practical File … 5 marks

PAPER I- THEORY: 70 Marks

S. NO. UNIT TOTAL WEIGHTAGE

1. Basic concepts of Artificial Intelligence 08 Marks
Introduction and State of Art of AI, Natural Language
2. 08 Marks
Processing (NLP), and Potential use of AI
3. Mathematics for AI 12 Marks
4. Data Visualization 16 Marks
5. Theoretical and Practical Aspects of Data Processing 08 Marks
6. Data Modelling, Simple Linear Regression 12 Marks
7. Ethical Practices in AI 06 Marks
TOTAL 70 Marks

ISC Examination Year 2027

1
 PAPER I – THEORY – 70 Marks Alexa),Internet searches such as navigation
searches (related to familiar brands and
1. Basic concepts of Artificial Intelligence platforms, e.g., LinkedIn, YouTube),
(i) Artificial Intelligence informational searches (for learning and
understanding), transactional searches (for
Definition, Evolution, Applications in
purchasing, signing up for services, or
different fields, commonly used AI
downloading apps), investigative searches
applications, benefits – decision making,
(e.g., top-rated web series, movies), and
remote patient monitoring, analysis of data,
voice searches.
solving complex problems, etc.
(iii) Potential use of AI
(ii) Role of data and information, evolution
computing Use of AI in various domains in Word
Processing System like Smart Phones, Web-
Types of data, identification, acquiring and
based Auction sites, Scanner machines, e-
exploring the data, binary logic system,
commerce platforms and social networking
conditional gates, deterministic and
sites (brief explanation).
probabilistic nature of real- life problems
with appropriate examples. (iv) AI and Society
(iii) Overview of Decision making Social benefits of AI: Healthcare
(enhancement in diagnosis treatment plans
Decision making in machines/computers;
and patient care), Transportation
Cyber security in computing and machine
(Autonomous vehicles and transport
intelligence.
management system), Disaster Prediction
(iv) Components of AI project framework (Early warning system and response
Problem scoping, data acquisition, data management) and Agriculture (Precision
exploration, modeling and evaluation (in farming, crop monitoring and yield
brief) prediction).
(v) Overview of Data representation and 3. Mathematics for AI
programming in Python
(i) Matrices
Datatypes, variables, operators,
conditional statements, control statements, Introduction to Matrices, Types of Matrices,
functions. Matrix Operations (Addition, subtraction,
multiplication, transpose).
2. Introduction and State of Art of AI, Natural (ii) Vectors and its applications
Language Processing (NLP), and Potential
Vector arithmetic
use of AI
(iii) Set Theory
(i) Brief History and Primary elements of AI
Introduction to data table joins, Context
Definition of Machine Learning (ML) and
setting, Set Theory and Relational Algebra,
Deep Learning (DL)/Neural Networks.
Set operations.
Application of ML and DL: Image
recognition/processing and Computer (iv) Simple Statistical Concepts
vision, Speech recognition, Information Measures of Central Tendency (Mean,
Retrieval (IR) through Search Engine, etc. Median, Mode), Variance and Standard
(ii) Domain of Natural Language Processing Deviation.
(NLP)
4. Data Visualization
Text understanding, Text generation,
Language translation (e.g., Google (i) Data Visualization using Python
translate), Question answering (Chatbots), Programming.
Dialogue systems (e.g., Siri and

ISC Examination Year 2027

2
 Using matplotlib and seaborn in Python, environmental , importance of AI ethics –the
and Excel charts; Handling missing values, effects of designing technology to replicate
outliers, and inconsistencies in data using human life.
Python's pandas library and Excel's data
cleaning features. PAPER II – PRACTICALS-30 Marks
(ii) Data Visualization using Statistical Graphs The practical paper of three hours duration will be
Types of Graphs-Bar Graph, Histogram, evaluated internally by the school. The paper shall
Scatter plot, Pie graph consist of three problem statements from which a
(iii) Introduction to Dimensionality of Data. candidate has to attempt any one problem statement.
Multi-dimensional data representation and The practical consists of two parts:
visualization using graphs. (1) Planning/Writing Session
(2) Examination Session
5. Theoretical and Practical Aspects of Data The total time to be spent on the Planning/Writing
Processing Session and the Examination session is three hours.
(i) Introduction to Data Cleaning. A maximum of 90 minutes is permitted for the
Data cleaning techniques with Pandas; Planning/Writing Session and 90 minutes for the
Handling duplicates and Inconsistent data. Examination session. Candidates are to be
permitted to proceed to the Examination Session
(ii) Exploring Kaggle Datasets. only after the 90 minutes of the Planning /
Creating and manipulating Data Frames Writing Session are over.
from Kaggle Datasets. Planning/Writing Session
(iii) Data Transformation and Standardization. The candidates will be required to prepare an
Methods for transforming and standardizing algorithm and a handwritten program to solve the
data for analysis. problem.
Examination Session
6. Data Modeling, Simple Linear Regression
The program handed in at the end of the
(i) Introduction to Data Modeling. Planning/Writing session shall be returned to the
A brief understanding of Types of Data candidates. The candidates will be required to do and
Models (Dimensional, relational and entity- execute the program individually on the computer,
relational). hardware and show execution to the examiner. A
printout of the program listing, including output
(ii) Regression analysis.
should be attached to the answer script containing the
Working of Regression (Dependent and handwritten program and hardware results. This
Independent Variables), Types of should be returned to the examiner. The program
Regression (in brief). should be sufficiently documented so that the
(iii) Linear Regression Equation. material required, circuit diagram/block diagram,
algorithm, representation and development process
Least Square Regression Line, Properties of
is clear from reading the program. Large differences
Linear Regression, Regression coefficient,
between the planned program and the printout will
Types of Linear Regression (in brief).
result in loss of marks.
(iv) Solving Linear Equations.
Teachers should maintain a record of all the
Applications of Linear Equations in various assignments done as part of the practical work
contexts. throughout the year and give it due credit at the time
of cumulative evaluation at the end of the year.
7. Ethical Practices in AI Students are expected to do a minimum of twenty
AI code of ethics- avoiding bias, ensuring assignments for the year and ONE project based on
privacy of users and their data, and mitigating the syllabus.

ISC Examination Year 2027

3
LIST OF SUGGESTEDASSIGNMENTS: Continuous Evaluation
1. How would you use Matplotlib to create a line Candidates will be required to submit a work file
plot for visualizing the trend of a dataset over containing the practical work related to assignments
time? Write a Python program demonstrating done during the year and ONE project.
this. Programming assignments done 10
2. Using Seaborn, how can you create a scatter throughout the year marks
plot to analyze the relationship between two
Project Work (based on any topic from 5
variables in a dataset? Provide a Python
the syllabus) marks
program illustrating this.
3. Explain how to utilize Matplotlib to generate a Proposed Guidelines for Marking
bar plot for comparing the distribution of
The actual grading will be done by the teacher based
categories in a dataset. Write a Python program
on his/her judgment. One possible way: divide the
for this purpose
outcome for each criterion into one of 3 groups:
4. Write a Python program demonstrating how to excellent, good, poor/unacceptable, then use
identify outliers in a dataset and handle them numeric values for each grade and add to get the
using outlier detection techniques. total.
5. Describe how to create a box plot using Seaborn
Evaluation will be done as follows:
to visualize the distribution of a numerical
variable. Provide a Python program to illustrate Programming Assignments:10 Marks
this.
Criteria Class design Documentation
6. How can you create a pair plot using Seaborn to (Total10 - Execution Practical File
visualize pairwise relationships between marks) (4 marks) (6 marks)
variables in a dataset? Write a Python program Excellent 4 6
demonstrating this.
7. Write a Python program using NumPy to Good 3 4
simulate coin flips and calculate the probability Poor 1 2
of getting heads.
8. Demonstrate how to generate random samples Project Work: 5 marks
from a normal distribution using NumPy and
Criteria Knowledge Functiona Presentation
visualize the distribution using Matplotlib.
(Total and lity and (1 marks)
Provide a Python program for this task.
5 marks) Understand Performa
9. Write a Python program to create a Pandas ing nce
DataFrame from a Kaggle dataset. (1 marks) (3 marks)
10. Perform a hypothesis test to determine if there Excellent 1 3 1
is a significant difference in temperature Good 1 2 1
between two cities. Write a Python script to Poor 1 1 1
conduct the test and interpret the results.
NOTE: This list is indicative only. Teachers and Terminal Evaluation
students should use their imagination to create
innovative and original assignments. Solution to Problem Statement on Hands- 15 marks
On/ Programming
EVALUATION OF PROGRAMMING
ASSIGNMENTS Marks should be given for choice of algorithm and
implementation strategy, documentation, correct
Marks (out of 30) should be distributed as given output on known inputs mentioned in the question
below. paper, correct output for unknown inputs available
only to the examiner.

ISC Examination Year 2027

4
 CLASS XII

There will be two papers in the subject: Paper II: Practical - 3 hours ... 15 marks
Paper I: Theory - 3 hours ... 70 marks Practical File … 15 marks

PAPER I- THEORY: 70 Marks

S. NO. UNIT TOTAL WEIGHTAGE

1. Applications of AI 10 Marks
2. Different paradigms of AI: Neural networks, Machine 15 Marks
learning, Deep learning.
3. Practical Implications of ANN 15 Marks
4. Practical Implications of Machine Learning (ML) 20 Marks
5. Introduction to Computer Vision (CV) 10 Marks
TOTAL 70 Marks

5
ISC Examination Year 2027
 PAPER I – THEORY – 70 Marks
Note: Key concepts of Class XI need to be revised like size, location, and number of bedrooms
as a prerequisite. using a decision tree.
(iii) Deep Learning (DL):
1. Applications of AI
About Deep Learning, Neural Network
(i) Natural Language Processing (NLP): utilized for image recognition, pattern
Named Entity Recognition (NER) recognition, and/or computer vision.
Identify and classify names of people, Example- Classifying images into different
organizations, and locations in a news categories (e.g., cat, dog, bird) using a
article, Text Summarization using Convolution Neural Network (CNN).
Extraction Methods: Summarize a long
passage by extracting the most important 3. Practical Implications of ANN
sentences or phrases, Spam Email Detection
(i) Data Preparation
using Machine Learning: Build a model to
classify emails as spam or non-spam based Use NumPy for numerical computations and
on their content and features. Pandas for data manipulation.
Employ Scikit-learn for data preprocessing
(ii) Statistics and Probability
tasks like scaling, encoding categorical
Understanding data trends using mean, variables, and splitting data into training
median, mode, range, variance, and and testing sets.
standard deviation; Probability Basics:
Introducing probability as a measure of (ii) Building the Neural Network
uncertainty, along with common Choose Keras or TensorFlow: Keras is easy
distributions like uniform, normal, and to use and runs on top of TensorFlow,
binomial. making it beginner friendly.
(iii) Hypothesis Testing (iii) Training the Neural Network
Understanding null and alternative Fit your data to the neural network model
hypotheses, Type I and Type II errors, and using the fit() function, specifying the
their relevance in AI experiments, number of epochs and batch size.
Regression Analysis: Introducing simple
(iv) Model Evaluation
linear regression for prediction and
correlation coefficient for measuring Evaluate the model's performance on the test
relationships between variables. data using metrics like accuracy, precision,
and recall.
2. Different paradigms of AI: Neural Networks, Visualize performance metrics using
Machine Learning, Deep Learning libraries like Matplotlib to gain insights.

(i) Artificial Neural Networks (ANN) (v) Tuning

About Artificial Neural Networks, nodes Utilize techniques like grid search or
(neurons), Input layer, Hidden layer, Output random search to tune hyper parameters
layer, Forward and Backward propagation such as learning rate, batch size, and
in ANN, ANN model, Multi-layer perceptron number of hidden layers and neurons.
(MLP), Example- Recognizing handwritten
digits using a multi-layer perceptron (MLP).
(ii) Machine Learning (ML) 4. Practical Implications of Machine Learning
Brief understanding of Machine Learning; (ML)
An AI paradigm enabling computers to learn (i) Predictive Maintenance
from data and make predictions or decisions
Using historical data from machines to
without explicit programming. Example-
predict equipment failure or require
Predicting house prices based on features

6
ISC Examination Year 2027
 maintenance, helping to minimize downtime PAPER II – PRACTICALS - 30 Marks
and optimize maintenance schedules.
The practical paper of three hours’ duration will be
(ii) Fraud Detection evaluated by the Visiting Examiner appointed
Credit card fraud or identity theft, enabling locally and approved by CISCE.
financial institutions to take proactive
measures to prevent losses. The paper shall consist of three problem statements
/problems from which a candidate has to attempt any
(iii) Medical Diagnosis one. The practical consists of two parts:
Analysing Medical Imaging, Genetic (1) Planning/ Writing Session
Information, and Patient records for (2) Examination Session
diagnosis.
The total time to be spent on the Planning/Writing
5. Introduction to Computer Vision (CV) Session and the Examination session is three hours.
A maximum of 90 minutes is permitted for the
(i) Introduction to CV, Application of Open CV Planning/Writing Session and 90 minutes for the
in real world scenario Examination session.
Image Classification, Object Detection, Candidates are to be permitted to proceed to the
Facial Recognition, Color Detection, Object Examination Session only after the 90 minutes of
Tracking (brief explanation with examples). the Planning / Writing Session are over.
(ii) Image Classification Planning/Writing Session
Implementing a simple image classifier to The candidates will be required to prepare an
distinguish between different categories of algorithm and a handwritten program to solve the
objects, such as cats vs. dogs or fruits vs. problem.
vegetables.
Examination Session
(iii) Object Detection
The program handed in at the end of the
Identification and localization of specific Planning/Writing session shall be returned to the
objects within an image, such as detecting candidates. The candidates will be required to code
cars in a street scene or detecting faces in a and execute the program individually on the
photograph. computer, hardware and show execution to the
(iv) Facial Recognition Visiting Examiner. A printout of the program listing
Building a facial recognition application including output results should be attached to the
that can recognize and identify faces in answer script containing the handwritten program
images or videos, and optionally, associate and hardware results. This should be returned to the
them with known individuals. examiner. The program should be sufficiently
documented so that the material required, circuit
(v) Color Detection diagram/block diagram, algorithm, representation
Identification and classification of objects and development process is clear from reading the
based on their color properties, such as program. Large differences between the planned
sorting colored objects on a conveyor belt or program and the printout will result in loss of marks.
detecting ripe fruits in agriculture.
Teachers should maintain a record of all the
(vi) Object Tracking assignments done as part of the practical work
Track the movement of objects in a video throughout the year and give it due credit at the time
sequence over time, such as tracking the of cumulative evaluation at the end of the year.
trajectory of a ball in sports footage or Students are expected to do a minimum of twenty-
monitoring vehicles in traffic surveillance. five assignments for the year.
The details are as follows:
Details of Assignments to be done during the year

7
ISC Examination Year 2027
 10. Introduction to CPU, GPU, TPU and DPU and
Broad Area Number of
their distinction for individual
Assignments
projects/research in AI.
Practical 15
(Programming) NOTE: This list is indicative only. Teachers and
students should use their imagination to create
Hands-on (Case Study) 10 innovative and original assignments.

TOTAL 25 EVALUATION OF PROGRAMMING

ASSIGNMENTS
Marks (out of 30) should be distributed as given
LIST OF SUGGESTED ASSIGNMENTS: below.
Some sample problems are given below as examples. Continuous Evaluation
The problems are of varying levels of difficulty:
Candidates will be required to submit a work file
1. Tokenization and Text Preprocessing:
containing the practical work related to assignments
Demonstrate tokenization techniques using
done during the year.
NLTK or spaCy in Python to break down
sentences into individual words or tokens. Programming assignments done 10
2. Named Entity Recognition (NER): Implement throughout the year (Internal marks
NER using spaCy to identify and classify Evaluation)
named entities such as persons, organizations, Programming assignments done 5
and locations in text data. throughout the year (Visiting Examiner) marks
3. Probability Distributions: Explore different
probability distributions (e.g., normal,
binomial, Poisson) using Python libraries like Proposed Guidelines for Marking
NumPy and visualize their properties using The actual grading will be done by the teacher based
Matplotlib. on his/her judgment. One possible way: divide the
4. Hypothesis Testing: Perform hypothesis testing outcome for each criterion into one of 3 groups:
(e.g., t-test, chi-square test) on sample data to excellent, good, poor/unacceptable, then use
make inferences about population parameters. numeric values for each grade and add to get the
5. Create a Python script to train a linear total.
regression model using the NumPy library to Evaluation will be done as follows:
predict a car's fuel efficiency (in miles per
gallon) based on its engine size (in liters). Programming Assignments: 10 Marks
6. Write a Python script to train a linear regression Criteria Class design Documentation
model using the NumPy library to predict a (Total 10 - Execution Practical File
student's final exam score based on their study marks) (4 marks) (6 marks)
hours. Excellent 4 6
7. Implementing Forward and Backward Good 3 4
Propagation: Build a simple neural network
from scratch in Python to understand the Poor 1 2
concepts of forward and backward propagation.
8. Introduction to Hidden Layers: Extend the basic
neural network by adding hidden layers and Terminal Evaluation
activation functions to demonstrate the role of
hidden layers in learning complex patterns. Solution to Problem Statement on 15 marks
9. Generate attendance monitoring system to track Hands-On/ Programming
the attendance of the students using Open-CV.

8
ISC Examination Year 2027
Marks should be given for choice of algorithm and II. Software Requirements:
implementation strategy, documentation, correct Operating System: Windows 10 or Linux
output on known inputs mentioned in the question (Ubuntu recommended for AI applications)
paper, correct output for unknown inputs available for flexibility in using open-source libraries.
only to the examiner. Programming Tools:
Note: Algorithm should be expressed clearly using  Python (latest version): Primary
any standard scheme such as a pseudo code. programming language for AI
applications.
LABORATORY REQUIREMENTS  IDEs:
- Jupyter Notebook or Google Colab
(For a class of 30 students)
(for cloud-based Python coding).
I. Hardware Requirements: - PyCharm or VS Code for offline
Group of 2 Students: 1 Desktop along with programming.
Webcam  AI Libraries:
- NumPy, Pandas, and Matplotlib: For
1. Webcam Detailed Description:
data manipulation and visualization.
Video Capture Resolution 1080p - Scikit-learn: For machine learning
Maximum Focal Length 1080 algorithms.
Maximum Aperture 2 f - TensorFlow or PyTorch: For deep
Reference: Logitech Brio 100 Full HD learning and neural networks.
Webcam - OpenCV: For computer vision
2. Desktop Configuration projects.
- NLTK or spaCy: For natural language
Detailed Description:
processing (NLP) projects.
Dell Optiplex 3000 Desktop:
 Database Tools: MySQL or SQLite for
12th Generation Intel Core I5-12500
storing and managing datasets.
Processor Intel B660 Chipset
 Cloud Platforms:
2GB NVidai Dedicated Graphics
- Google Colab or Amazon Web
ENERGY STAR Qualified
Services (AWS) for running large AI
8GB, DDR4, Non-ECC Ram 3200 Mhz
models.
M.2 512GB Gen 4 PCIe NVMe Solid
- GitHub for version control and
State Drive
collaborative projects.
Power Cord 1M for India Dell USB
Keyboard & Optical Mouse III. AI Kits (Optional but Beneficial):
Windows 11 Pro OEM with License  Raspberry Pi AI Kits: For experimenting
Dell 24” Display with AI on hardware devices.
Wifi Card + HDMI 1 Mtr. Cable
 Robotics kits: For AI integration into
3 Years onsite Warranty robotics projects.
Micro-Tech UPS - 600 VA (15 mins)

9
ISC Examination Year 2027
 SAMPLE TABLE FOR PRACTICAL WORK
Assessment of Assessment of the Practical Examination TOTAL MARKS
Unique Practical File (To be evaluated by the Visiting Examiner only) (Total Marks are to
Identification be added and
S. No. Number (Unique Internal Visiting Algorithm Python Program Hard Output entered by the
ID) of the candidate Evaluation Examiner with internal Copy Visiting Examiner)
10 Marks 5 Marks Documentation (printout)
3 Marks 7 Marks 2 Marks 3 Marks 30 Marks
1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Name of the Visiting Examiner:_________________________________

Signature: _______________________________
Date:___________________________________

10
ISC Examination Year 2027