Data Science Essentials 101
[Link]@[Link]
Z5JNEQFHRL
The mathematics and statistics underlying
the technology
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Learning Objectives
Upon completion of this module, you will be able to:
Describe how math and statistics are leveraged in data science.
Explore Bayes Rule and Naive Bayes by delving deeper into Probability, using a
Covid-19 scenario.
Gain insights
[Link]@[Link]
Z5JNEQFHRL on matrices and linear algebra, and how it is used in data
science, through a simple recommendation system, image representation
and dimension reduction.
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Learning Outcomes
You should now be able to:
Comprehend the importance of math and stats in data science.
Gain a more precise understanding of how linear algebra is employed in data
representation.
Demonstrate
[Link]@[Link] the utilization of calculus in optimizing solutions.
Z5JNEQFHRL
Enumerate on how probability is used for decision making.
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Agenda
In this session, we will discuss:
The need for math in Data Science
Probability and Descriptive Statistics
Matrices
[Link]@[Link]
Z5JNEQFHRL and Linear Algebra
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
The need for math in Data Science
Sales
Machine
Learning &
Computer Deep Learning
Science & Math & Statistics
Technology
[Link]@[Link] Data
Z5JNEQFHRL Science
Software
Development Research Price
Sales = 500-20*Price
Domain/Business
Knowledge
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Probability & Descriptive Statistics
[Link]@[Link]
Z5JNEQFHRL
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Bayes’ Rule
“Determine probability of a hypothesis based on prior knowledge and new evidence.”
Eg: Shopping patterns of 100 people in a retail store Probability of a new person to enter the
food section?
Probability (Food) = 0.6
Food
Probability of a person to enter the clothing
60 section?
Probability (Clothing) = 0.4
[Link]@[Link]
Z5JNEQFHRL
Probability of a person in the shirt section,
100 given, he is in the clothing section?
20
Probability (Shirt | Clothing) = 0.5
Shirts
40
Clothing
20 Video Thumbnail
DO NOT REMOVE
Pants
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Descriptive Statistics
“Describing features of a dataset by generating summaries about data samples ”
Average Return = Mean
[Link]@[Link]
Z5JNEQFHRL
Risk = Standard Deviation
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Descriptive Statistics
“Describing features of a dataset by generating summaries about data samples ”
[Link]@[Link]
Z5JNEQFHRL
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
[Link]@[Link]
Z5JNEQFHRL Matrices & Linear Algebra
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
A simple recommendation system
“Personalized recommendations to users based on their past behavior & preferences”
Movies -> Items
Approach - 1
item1 item2 item3 item4 item5
You watched:
user1 5 2 1 4 1
Viewers -> Users
user2 4 ? 1 3 3
[Link]@[Link]
So I’ll recommend: user3
Z5JNEQFHRL 3 3 2 ? 4
user4 2 1 2 2 5
Approach - 2 user5 5 3 ? 4 3
You are between age 20 to 30, prefer fantasy, action & Ratings
science fiction
Gaps
So I’ll recommend: Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
A simple recommendation system
“Personalized recommendations to users based on their past behavior & preferences”
User i has an average rating ai
bj Item i has an average rating bi
item1 item2 item3 item4 item5
ai x bj
[Link]@[Link]
Z5JNEQFHRLuser1 5 2 1 4 1
user2 4 ? 1 3 3
user3 3 3 2 ? 4
ai
user4 2 1 2 2 5
user5 5 3 ? 4 3
User-Item Matrix
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Image Representation
“Convert an image into an array of numbers and each pixel has an intensity number”
010 146 025
057 071 082
[Link]@[Link]
Z5JNEQFHRL
012 098 099
128 x 128 array 128 x 128 array
0 255
Black White
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Image Representation
“Convert an image into an array of numbers and each pixel has an intensity number”
163 164 172
Red
111 162 118
035 093 025
Green 116 111 152
[Link]@[Link] 041 042 051
Z5JNEQFHRL 032 081 082
071 081 025
011 012 072
Blue
128 x 128 array 061 192 142
0 255
No Red Deep Red
No Green Deep Green Video Thumbnail
DO NOT REMOVE
No Blue This file is meant for personal use by [Link]@[Link] only.
Deep Blue
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Dimension Reduction
“Increasing the interpretability of data while preserving the maximum amount of information”
[Link]@[Link]
Z5JNEQFHRL
Selecting pixels that have maximum variance = maximum information
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
Summary
Here’s a brief recap:
Math and Statistics are the foundations that help translate business problems into
solutions.
Through basic probability and descriptive statistics, we can derive answers to our
questions by simple observation and inferences.
The underlying principle behind simple recommendation systems is the usage of
matrices
and arrays to map users and their preferences.
[Link]@[Link]
Z5JNEQFHRL
Linear algebra and matrices also play a pivotal role in representing grayscale and
color images into arrays, which is used in image processing and other applications.
Foundational math and statistics also finds its application in dimensionality reduction
- enabling interpretation of data while preserving only the important information.
Video Thumbnail
DO NOT REMOVE
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
H appy Learning !
[Link]@[Link]
Z5JNEQFHRL
This file is meant for personal use by [Link]@[Link] only.
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited 17
Sharing or publishing the contents in part or full is liable for legal action.
Proprietary content. © Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.