2 3

1 Sustainable Development Goals

What is Project Cycle

Project Cycle is a step-by-step process to solve problems using proven Sustainable Development: To Develop for the present without exploiting the
scientific methods and drawing inferences about them. resources of the future.
17 goals announced by United Nations.
Components of Project Cycle Aim to achieve them by 2030.
Pledge taken by all the member nations of the UN.
Components of the project cycle are the steps that contribute to completing
The Sustainable Development Goals (SDGs), also known as the Global Goals,
the Project. The Components of AI Project Cycle are:-
were adopted by all United Nations Member States in 2015 as a universal call to
action to end poverty, protect the planet and ensure that all people enjoy peace
Problem Scoping - Understanding the problem
and prosperity
Data Acquisition - Collecting accurate and reliable data
Data Exploration - Arranging the data uniformly

* Image is the property of individual organisation and is used for reference only.
Modelling - Creating Models from the data
Evaluation - Evaluating the project

Problem Scoping

Problem Scoping refers to

understanding a problem,
finding out various factors which
affect the problem, define the
goal or aim of the project.
 4 5

2 4 W's of Problem Scoping The Stakeholder Who

The 4W’s of Problem Scoping are Who, What, Where, and Why. This Ws Have a problem Issue/Problem What
helps in identifying and understanding the problem in a better and efficient
manner. When/While Context/Situation/Location Where

Who - “Who” part helps us in comprehending and categorizing who all Ideal Solution How the Solution will help Stakeholders Why
are affected directly and indirectly with the problem and who are called
[Problem Statement Template]
the Stake Holders
Data Acquisition
What - “What” part helps us in understanding and identifying the nature 2 Types of Data Sets
of the problem and under this block, you also gather evidence to prove The process of collecting accurate and reliable data to work with.
that the problem you have selected exists.

Where - "Where” does the problem arise, situation, context, and location.

Why - “Why” is the given problem worth solving.

4 Problem Statement Template

Data features ⚆_⚆ → Refer to the type of data you want to collect.
Ex: salary amount, increment percentage, increment period, bonus, etc.
The Problem Statement Template helps us to summarize all the key points into
one single template. Big Data ⚆_⚆

So that in the future, whenever there is a need to look back at the basis of the It includes data with sizes that exceed the
problem, we can take a look at the Problem Statement Template and capacity of traditional software to process
within an acceptable time and value.
understand its key elements of it.
The main focus is on unstructured type of Amount of
Types of
Speed of
Have a look at Problem Statement Template. data. Data
Data
Produced
Data
Produced Produced
 6 7

2 Data Sources Data Exploration

Web Scraping Sensors Data Exploration is the process of arranging the gathered data uniformly for a
better understanding. Data can be arranged in the form of a table, plotting a
Web Scraping means collecting data Sensors are very Important but
chart, or making a database.
from web using some technologies. very simple to understand.
We use it for monitoring prices, Sensors are the part of IoT To analyse the data, you need to visualise it in some user-friendly format so
news and etc. (Internet of things) that you can:
Quickly get a sense of the trends, relationships and patterns
Example: Web Scrapping. using Sensers collect the physical
Define strategy for which model to use at a later stage
beautiful soup in python. data and detect the changes.
Communicate the same to others effectively

Cameras Observations 1 Data Visualisation tools

Camera captures the visual When we observe something The tools used to visualise the acquired data are known as data visualisation
information and then that information carefully we get some information or exploration tools.
which is called image is used as a For ex: Scientists Observe
source of data. creatures to study them. Few data visualisation tools are: Google Charts, Tableau, Fusion Charts,
Cameras are used to capture raw Highcharts
Observations is a time consuming
visual data. data source.

API Surveys
Application Programming interface. The survey is a method of
gathering specific information
API is a messenger which takes
from a sample of people.
requests and tells the system about
requests and gives the response. Example, a census survey for
analyzing the population.
Ex: Twitter API, Google Search API
 8 9

AI Modelling → 2 ways/Approaches→→ Rule Based Approach

Learning Based Approach

1 Learning Based Approach

Modelling is the process in which different models based on the visualized The learning-Based Approach is based on a Machine learning experience with
data can be created and even checked for the advantages and disadvantages the data fed.
of the model.
Machine Learning (ML)
1 Rule Based Approach Machine learning is a subset of artificial intelligence (AI) that provides
machines the ability to learn automatically and improve from experience
Rule Based Approach Refers to the AI modelling where the relationship
without being programmed for it.
or patterns in data are defined by the developer.
That means the machine works on the rules and information given by the
developer and performs the task accordingly.

Ex: You trained your model with 100 images of apples and bananas. Now If you test
it by showing an apple, it will figure out and tell if it's an apple or not. Here Labeled
images of apple and banana were fed, due to which the model could detect the fruit.

*Labeled Images: Simply, when the model is told about what the image is.
Data Sets

Dataset is a collection of related sets of information that is composed of Types of Machine Learning
separate elements but can be manipulated by a computer as a unit.

3 types of Machine Learning:-

Training Data – A subset required
Supervised Learning
to train the model
Unsupervised Learning
Testing Data – A subset required Semi-supervised or
while testing the trained the model
Reinforcement Learning
 10 11

Supervised learning → 2 Categories →→ Classification

Regression
Unsupervised Learning

Supervised learning is where a computer algorithm is trained on input data that In terms of machine learning, unsupervised learning is in which a system learns
has been labeled for a particular output. through data sets created on its own. In this, the training is not labeled.

→ Classification Important Points:

An unsupervised learning model works on unlabelled dataset.
Here, Data is categorized under different labels
This means that the data which is fed to the machine is random and there
according to some parameters given in the input
is a possibility that the person who is training the model does not have
and then the labels are predicted for the data.
any information regarding it.
Example: To predict which of them is apple The unsupervised learning models are used to identify relationships,
and banana. patterns and trends out of the data which is fed into it

It helps the user in understanding → What the data is about

→ What are the major features identified by
→ Regression the machine
Regression is a type of supervised learning Example: Suppose a boy sees someone performing tricks with a ball, so he
which is used to predict continuous value. also learnt the tricks by himself. This is what we call unsupervised learning.

Example: To predict your next salary, put in the

data of your previous salary, any increments, etc.,
→ Clustering
train the model. Its an algorithm which can cluster the unknown
data according to the patterns or trends identified
Example: Weather Prediction using past data. out of it
Here, the data which has been fed to the machine is continuous. The patterns observed can be known to the
developer or it can be unique.

Note: Classification ≈ Division, Clustering ≈ Grouping

 12 13

→ Dimensionality Reduction: Training vs Testing Data

We can visualize up to 3-Dimensions only.
Training Set Testing Set
To reduce the dimensions and still be able to make Base

sense of the data, we use Dimensionality Reduction. Use Used for Training the Model Used for Testing the Model after it is trained
The ball in our hand is 3-Dimensions. But if we click
Is a lot bigger than testing data and It is smaller than Training Set and
its picture, the data transform to 2-D. Size
constitutes about 70% to 80% constitutes about 20% to 30%

Note: Classification ≈ Division, Clustering ≈ Grouping

Evaluation
Reinforcement Learning
Evaluation is the process of understanding the reliability of any AI model, based on
Learning through feedback or trial and error method is called Reinforcement outputs by feeding the test data into the model and comparing it with actual answers.
Learning. There can be different Evaluation techniques, depending on the type and
purpose of the model.
The system works on Reward or Penalty policy. In this an agent performs an
action positive or negative, in the environment which is taken as input from the More about Evaluation at - Unit 7 Evaluation
system, then the system changes the state in the environment and the agent is
provided with a reward or penalty.
MORE from AIFORKIDS
Example: A very good example of
these is Vending machines.