AI PORTFOLIO ACTIVITIES

ACTIVITY - 1
AI PROJECT CYCLE
PROBLEM SCOPING
Step1: Select the theme

Step2: Scope the list of problems

Step3: Fill the 4W’s Problem Canvas

1. Who canvas – Who is having the problem?

2. What Canvas – What is the nature of their problem?
3. Where Canvas - What is the nature of their problem?
4. Why Canvas – Why do you think it is a problem worth solving?

After filling the 4Ws problem canvas, we need to summarise all the canvas into one template.
 ACTIVITY - 2
AI PROJECT CYCLE
DATA EXPLORATION

BASED ON DATA VISUALIZATION CATALOGUE

How to Use Data Visualization Catalogue Online?
Open a web browser and type the following URL: https://datavizcatalogue.com/ in the address bar and press
Enter. You will get the following screen:
Here, you can see various types of graphical representations like arc diagram, area graph, etc.
By clicking the icons, look at the various new ways of data Visualization and identify the ones which interest
you the most. Write down at least 5 new data Visualization techniques in the format given below:
For example:-
DATA VISUALISATION TECHNIQUES 1
Name of the Bar Graph
Representation
One-line Description The classic Bar Chart uses either horizontal or vertical bars to
show discrete, numerical comparisons across categories.
How to draw it?

Suitable for which data Bar charts are great when we want to track the development of one
type? or two variables over time.
Functions Comparison and Functions
Similar charts/graphs

DATA VISUALISATION TECHNIQUES 1

Name of the
Representation

One-line Description

How to draw it?

Suitable for which data

type?

Functions

Similar charts/graphs
 ACTIVITY - 3
AI PROJECT CYCLE
DATA MODELLING

PIXEL IT ACTIVITY

Purpose

The main purpose of this activity is to understand how the computer classifies the images as well as how a
computer reads them. The graphics or images created in computers are pixel-based images. 1 pixel means 1
dot. So in this activity, you will understand how the images will be processed, classified and how computers
see them. This is one of machine learning approach used in CV (Computer Vision) applications.

We are performing to understand the learning based approach for modelling.

Follow this steps:

Step1: Cut the page or draw 6 x 6 matrix square boxes on the sheet.

Step2: Write an upper case alphabet on the matrix with the equal height of the matrix. It should cover all the
boxes starting from top to bottom.

Step3: Now apply color to the boxes on which lines of that alphabet have fallen.

Step4: Color the boxes.

Step5: Now cut the horizontal stripes of the matrix in this manner like 1-2,2-3,3-4,4-5,5-6 and 6-7.

Step6: Cut the horizontal stripes.

Step7: Now, paste all these stripes together to form a single paper string. Make sure that the last block should
neither be over the first block of the next line nor should there be any gap in between the first and the last
blocks.

Step8: Take the value '1' of the box or cell color which is red and '0' where the box or cell colour is empty as
shown.
Step9: Take the second paper and draw the letter 'R' again randomly as done in steps 2&3, as shown in the
figure. (You can take any color to fill the cells or boxes.)

Step10: Cut the strips as done previously in step 6 and arrange the strips as done in step 7.
Step11: Take values '1' or '0' as done in step 8.
Now take both letters' values and add them in a sequence as shown.

Step12: Now take the sum value of the first and second letter and place its value in table cells on the third
paper as shown in figure.
Fill any color where the table cell values have '1' or 2' value and leave those cells empty whose value is '0' as
shown in figure.

Here, you can see that the output is similar to the letter "R".

What we have learnt?

Pixel It activity is an example of how computers see images, process them and classify them. This kind of
Machine Learning approach is commonly used in Computer Vision related applications. Every image which is
fed to the computer is divided into pixels (which are the smallest unit of an image). The Computer analyses
each pixel and if it has to compare 2 pictures to check if they are similar or not, pixel-wise comparison takes
place. If pixels are identical, this means that the images are the same.

In this activity, you created your own intelligent model which can identify if an alphabet is similar to the data
trained or not. Here, you divided your alphabet into 36 blocks and then processed it. Also, to train the model,
you used datasets of the same alphabet but different handwritings. Now, when you go for testing it, the model
would see if the coloured blocks are aligned or not. If majority of the blocks are aligned altogether, there
exists maximum probability that the alphabet is the same. Otherwise the alphabet is different.

Learning outcome

This activity clarifies that:

 The images saved in the computer are divided into the pixels.
 Machine learning also uses the same approach for its applications.
 The computer analyses each pixel, whenever any comparison required it compares these pixels. If
pixels are identical, this means that the images are the same.

By the above activity, we have created an intelligent model to identify the alphabet is same or not. We have
divided the alphabet into 36 blocks and processed it. Trained the model to process the same alphabet but
different handwritings. Now when we testes it, the model would see if the colored blocks are aligned or not. If
the majority of the blocks are aligned altogether, there exists a maximum probability that the alphabet is the
same. Otherwise, the alphabet is different.

For better understanding of this activity you can watch the following video.
https://youtu.be/TcVCjetCC7Q

To understand Neural Network concepts you can watch the following videos.

https://www.youtube.com/watch?v=bUBRPFnmxqI
https://drive.google.com/file/d/1ll2wbXnJu8GsIRIQxi9Xc6YnuZt_TC6D/view?usp=share_link
https://drive.google.com/file/d/1Pi2LT1YRNagi_JD_8sycysOrDTDzoG9N/view?usp=share_link