GREENDALE SCHOOL

Affiliated to CBSE, New Delhi

PM Palem | Madhurawada | Visakhapatnam – 530041
----------------------------------------------------------------------------------------------------------------------------- ---

Subject: ARTIFICIAL INTELLIGENCE

2. ADVANCED CONCEPTS OF MODELING IN AI

I. Short answer type questions.
Q1. What is Machine learning?
Ans. Machine learning (ML) is a type of artificial intelligence (AI) that enables machines to
learn from data and improve over time. ML is used to create systems that can perform
tasks without being explicitly programmed.

Q2. Differentiate between classification and regression model.

Ans.
Classification
 Classification gives out discrete values.
 Given a group of data, this method helps group the data into different groups.
 In classification, the nature of the predicted data is unordered.
 The mapping function is used to map values to predefined classes. It attempts to
find a best fit line.
 Example include Decision tree, logistic regression.
 Classification is done by measuring the accuracy.
Regression :
 Regression gives continuous values.
 It uses the mapping function to map values to continuous output.
 Regression has ordered predicted data.
 Regression is done using the root mean square error method.
 It tries to extrapolate the graph to find/predict the values.
 Examples include Regression tree (Random forest), Linear regression

Q3. What are ANN’s?

Ans. Neural networks also known as artificial neural networks are computational networks
that are fundamental component of deep learning algorithms. Neural networks are
subfield of AI and the subset of machine learning. They are designed to mimic the
structure of human brain and inspired by how the human brain interprets and processes
information.

Q4. What happens in the output layer of a neural network

Ans. In the output layer of a neural network, the final predictions or results are produced
after processing data from the preceding layers.

Q5. Define CNN.

Ans. Convolutional neural network is a deep learning algorithm specifically design for
processing image data. It receives an input image, assign importance to various aspects
of the image, and extract important features. This helps differentiate one image from the
other.
Q6. Define association in unsupervised learning.
Ans. In unsupervised learning, "association" refers to finding common patterns and
relationships between different attributes or items within a dataset, without any predefined
labels or goals.

II. Long answer type questions.

Q1. Differentiate between the rule based approach and learning based approach.
Ans.
Following are the differences between the rule-based approach and the learning-based
approach:
Rule-based approach
 The machine follows the rules defined by
 the developer.
 AI is achieved through rule-based technique.
 It typically uses labelled data,
 It may require less training time.
Learning-based approach
 The machine learns on its own from the
 data.
 AI is achieved through learning technique.
 It can handle both labelled and unlabelled Data.
 It requires more training time.

Q2. What is supervised, unsupervised and reinforcement learning? Explain with

examples.
Ans.
1. Supervised Learning:
Concept:
This type of machine learning uses labeled data (where the input and output are known)
to train a model to predict or classify new, unseen data.
Examples:
Image Recognition: Training a model to recognize cats in images, where each image is
labeled as "cat" or "not cat".
2. Unsupervised Learning:
Concept:
This type of machine learning learns from unlabeled data, meaning the model doesn't
have pre-defined outputs or labels.
Examples:
Recommender Systems: Suggesting products or content to users based on their past
interactions (e.g., Netflix suggesting movies).
3. Reinforcement Learning:
Concept:
This type of machine learning involves an "agent" (the AI model) that learns to make
decisions in an environment to maximize a reward
Examples:
Game Playing: Training AI to play games like chess or Go, where the AI learns by playing
and receiving rewards for winning.

Q3. Explain neural networks. Also, describe the role of input, hidden and output layers in
a neural network.
Ans.Neural networks are computational models inspired by the human brain, used for
making predictions or classifications from data. They consist of interconnected nodes
(neurons) organized into layers: input, hidden, and output, each playing a crucial role in
processing information.
The different layers in ANN’s are:
Input Layer:
This layer receives the raw data or features that the network will process, acting as the
entry point for information.
Hidden Layer(s):
These layers perform complex computations on the input data, learning patterns and
features that are then passed to the next layer. Networks with multiple hidden layers are
known as deep neural networks.
Output Layer:
This layer produces the final prediction or classification based on the processed
information from the hidden layers.

Q4. Describe the following AI models and draw an appropriate diagram:

a) Regression
b) Clustering
Ans. a) Regression: Regression is a learning-based AI model used to predict continuous
numerical values. It is used to predict the behavior of one variable depending upon the
value of another variable.
A special function called mapping function helps us
predict future values based on the data we have.
For instance, we can use an employee’s past salary
data to train the AI model to predict the future
salary. The algorithm learns from this data and
creates a solid line on a graph that represents the
pattern in the salaries. In the given graph, the blue
dots represent the data values we have i.e. the
past salaries, and the solid line represents the
function that helps us predict future values.

b) Clustering: Clustering is an unsupervised machine learning technique where the model

groups a data set into different clusters or groups based on similarities in data points.
Unlike supervised learning, clustering does not require labelled data; the model identifies
patterns or trends on its own to categorise data points. The data in the model is usually
unlabelled or random.
Q5. Neural networks are said to be model the way how neurons in the human brain
behave. A similar system is mimicked by the AI machine to perform certain tasks. Explain
how neural networks work in any eye model and mention any three features of neural
networks
Ans.A neural network is a technique in Artificial Intelligence (AI) that instructs computers
to function data in a manner that is promoted by the human brain. It is a kind of machine
understanding process known as deep learning, that utilizes interconnected nodes or
neurons in a layered pattern that matches the human brain.
Neural network training is the technique of instructing a neural network to accomplish a
task. Neural networks understand by originally processing various large sets of labelled or
unlabelled data.
Three Features of Neural Networks:
1. Adaptability:
Neural networks can learn and adapt to new data and patterns, improving their
performance over time.
2. Parallel Processing:
They can process multiple inputs simultaneously, making them efficient for large datasets.
3. Non-Linearity:
They can model complex, non-linear relationships in data, allowing them to solve
problems that traditional algorithms struggle with.