CS3352 FOUNDATIONS OF DATA SCIENCE LTPC 3003

COURSE OBJECTIVES:
 To understand the data science fundamentals and process.
 To learn to describe the data for the data science process.
 To learn to describe the relationship between data.
 To utilize the Python libraries for Data Wrangling. To present and interpret data using
visualization libraries in Python

UNIT I INTRODUCTION 9
Data Science: Benefits and uses – facets of data - Data Science Process: Overview – Defining
research goals – Retrieving data – Data preparation - Exploratory Data analysis – build the
model– presenting findings and building applications - Data Mining - Data Warehousing – Basic
Statistical descriptions of Data
UNIT II DESCRIBING DATA 9
Types of Data - Types of Variables -Describing Data with Tables and Graphs –Describing Data
with Averages - Describing Variability - Normal Distributions and Standard (z) Scores
UNIT III DESCRIBING RELATIONSHIPS 9
Correlation –Scatter plots –correlation coefficient for quantitative data –computational formula
for correlation coefficient – Regression –regression line –least squares regression line – Standard
error of estimate – interpretation of r2 –multiple regression equations –regression towards the
mean
UNIT IV PYTHON LIBRARIES FOR DATA WRANGLING 9
Basics of Numpy arrays –aggregations –computations on arrays –comparisons, masks, boolean
logic – fancy indexing – structured arrays – Data manipulation with Pandas – data indexing and
selection – operating on data – missing data – Hierarchical indexing – combining datasets –
aggregation and grouping – pivot tables
UNIT V DATA VISUALIZATION 9
Importing Matplotlib – Line plots – Scatter plots – visualizing errors – density and contour plots
– Histograms – legends – colors – subplots – text and annotation – customization – three
dimensional plotting - Geographic Data with Basemap - Visualization with Seaborn.
COURSE OUTCOMES:
At the end of this course, the students will be able to:
CO1: Define the data science process
CO2: Understand different types of data description for data science process
CO3: Gain knowledge on relationships between data
CO4: Use the Python Libraries for Data Wrangling
CO5: Apply visualization Libraries in Python to interpret and explore data
TOTAL:45 PERIODS
TEXTBOOKS:
1. David Cielen, Arno D. B. Meysman, and Mohamed Ali, “Introducing Data Science”, Manning
Publications, 2016. (Unit I)
2. Robert S. Witte and John S. Witte, “Statistics”, Eleventh Edition, Wiley Publications, 2017.
(Units II and III)
3. Jake VanderPlas, “Python Data Science Handbook”, O’Reilly, 2016. (Units IV and V)

REFERENCE:
1. Allen B. Downey, “Think Stats: Exploratory Data Analysis in Python”, Green Tea Press,2014.
COURSE TITLE: FOUNDATION OF DATA SCIENCE CODE: CS3352
YEAR/SEMESTER: II / III

COURSE OBJECTIVES:
The students should be made to:

S.No Objectives
1 To understand the data science fundamentals and process.
2 To learn to describe the data for the data science process.
3 To learn to describe the relationship between data.
4 To utilize the python libraries for data wrangling.
5 To present and interpret data using visualization libraries in python.

COURSE OUTCOMES:
At the end of the course, the students should be able to:

CO’s Course Outcomes

1 Identify the data science process.
2 Explain different types of data description for data science process.
3 Recognize the relationship between data.
4 Apply the python libraries for data wrangling.
5 Apply visualization libraries in python to interpret the explore data.

MAPPING OF COURSE OUTCOMES AND PROGRAM OUTCOMES:

Course Program outcomes / Program Specific Outcomes

Outcomes 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1 - 2 1 2 - - - - - - - 1 1 -
2 - 2 - 2 1 - - - - - - 1 1 -
3 1 1 - 2 2 - - - - - - 1 2 -
4 2 1 2 1 2 - - - - - - 2 2 -
5 2 1 2 1 2 - - - - - - 2 2 -

GAP IN THE SYLLABUS

1. Big Data
2. Data Analytics

CONTENT BEYOND THE SYLLABUS

1. Big Data
2. Data Analytics
 Subject Handler HOD / IT

COURSE TITLE: DATA SCIENCE LAB CODE: CS3362

YEAR/SEMESTER: II / III

COURSE OBJECTIVES:
The students should be made to:

S.No Objectives
1 To understand the data science fundamentals and process.
2 To learn to describe the data for the data science process.
3 To learn to describe the relationship between data.
4 To utilize the python libraries for data wrangling.
5 To present and interpret data using visualization libraries in python.

COURSE OUTCOMES:
At the end of the course, the students should be able to:

CO’s Course Outcomes

1 Use python ibraries for data science
2 Perform various analytics on the standard data sets.
3 Interpret data using visualization packages in python.

MAPPING OF COURSE OUTCOMES AND PROGRAM OUTCOMES:

Course Program outcomes / Program Specific Outcomes

Outcomes 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1 - 2 1 2 - - - - - - - 1 1 -
2 - 2 - 2 1 - - - - - - 1 1 -
3 1 1 - 2 2 - - - - - - 1 2 -
4 2 1 2 1 2 - - - - - - 2 2 -
5 2 1 2 1 2 - - - - - - 2 2 -

CONTENT BEYOND THE SYLLABUS

1. Python program to convert csv file to text file, create and excel file and add a
column.

Subject Handler HOD / IT

Why use Python for big data & analytics?
When it comes to data analytics or data science, its sheer
complexity becomes a major concern. So much so that you might
think you need a specialized programming languages to handle such
tasks. Of course, there are indeed programming languages that
specialize in those fields—R, Scala, Julia, etc.
Yet, in most cases, data analytics professionals and data scientists
actually prefer Python over these other languages! There are
several reasons behind this popularity, let’s have a look.

(Compare Python to Java & Go.)

Ease of use
Python is relatively easy to learn and much less wordy compared to
other languages like Java. (This characteristic is known as being
Pythonic.) This simplicity lowers the barrier to enter Python as a new
programming language.

The best part is that the simplicity of the language does not affect
the functionality in any shape or form, and Python is always
powerful. Simply install the language, and you are ready to get
started. There are no complex configurations required, such as
setting up compilers.

Additionally, Python is not restricted to a particular programming

style. Therefore, users are free to choose a programming
methodology that they are comfortable with:

 Object oriented programming (OOP)

  Functional programming
 Etc.

Finally, the Python language consists of more readable code, which

in turn helps users easily understand codebases. Python can also act
as the gateway for big data and data science fields without having
to learn a new language.

(New to Python? Get to work with this Python starter kit.)

Licensing structure
Python is an open-source language managed by the non-profit
Python Software Foundation. This open-source nature allows Python
to be used in any project without fearing any interference from a
third party. Contrast this to other programming languages managed
or owned by commercial organizations, where a single decision can
cripple the usage of the language.
Data analytics projects can be complex and time-consuming
endeavors. Thus, the open-source nature of Python helps data
scientists and analysts confidently use it for the foreseeable future
of any kind of commercial or hobbyist project.

Active community
Python has an active and thriving community, members of which are
routinely and actively:

 Contributing to Python language development.

 Creating and maintaining thousands of packages that help
enhance and extend core Python functionality.

Thanks to this large community, there are many online

resources, GitHub/GitLab projects, forums, and communities
available to anyone. Furthermore, there is a high chance that you
will be able to find solutions to most issues by simply googling.
Python tools for Big Data and Analytics
So far, we have covered the general characteristics of Python that
make it an ideal choice for Big Data and Analytics. Now let’s dig a
bit deeper and explore the tools available in Python to cater to these
needs.

Python packages
Python comes with the Python Package Index (PIP), the open-source
repository that contains all the third-party packages available for
Python. This library consists of packages to help users in various
tasks, from simple tasks like JSON parsing to complete data
transformation, analytics, and visualizations packages.

Let’s look at some of these packages.

The Pandas data analytics library is one of the most prominent

packages out of these packages. Pandas have become the defector
standard for both novice and experienced users in data science,
engineering, and analytics fields. Furthermore, it can handle any
kind of tabular data sets, enabling users to explore, clean,
transform, filter, and process data.
Pandas is versatile enough to handle any data set with its ability to
import and export data to and from various sources easily.
Additionally, Pandas can be used to create plots, handle time-series
data and textual data. All these facts have led Pandas to become
one of the most powerful libraries available for data analytics.

(Explore our hands-on Pandas Guide.)

Next up: NumPy (Numerical Python) is there if you need a
package to handle scientific computing and mathematical functions.
NumPy provides a powerful multidimensional array object and a
whole bunch of routines to manipulate arrays such as mathematical,
statistical, basic linear algebra, sorting, logical functions, etc.
Due to the vectorized code, NumPy can process data faster than any
other library, which is highly beneficial when dealing with complex
data sets. The NumPy API is also used extensively to enhance
functionality in other packages, such as

 Pandas
 SciPy
 scikit-learn
 And more

(Start with these NumPy basics.)

Visualizing data is as important as analytics. In Python, we have
powerful packages such as matplotlib, seaborn, folium, and even
ploty that offer comprehensive visualization capabilities. For
example, the matplotlib package not only allows users to create
static visualization but also offers animated interactive visualization
with full control over all the aspects of the visualization. These
functions can be extended further by integrating third-party
packages like base map, cartopy, etc.
In addition to the packages mentioned above, there are numerous
other packages geared towards data analytics and processing, such
as Polaris, Desk, Vaex, PySpark, etc.

(Explore our Data Visualization Guide, which explores many of these

tools & packages.)
Anaconda
Anaconda Data Science Toolkit, or simply Anaconda, is an open-
source toolkit that aims to facilitate a complete data science
platform with a single installation. Everything is seamlessly
integrated with Anaconda, besides setting up a Python environment
and installing and managing packages separately.

Anaconda provides the conda package and environment

management system, built-in machine learning, data science, and
visualization libraries under a single software package. In addition to
Python support, Anaconda also supports R allowing users to utilize
the same familiar toolkit with a specialized data science
programming language.

On top of that, Anaconda Navigator users get a complete GUI to

manage packages, environments, and software without relying on
the command line. Anaconda supports various applications like
Spyder, JupyterLab, Datalore, etc., which can also be managed via
the GUI. All these features are offered for free with the Anaconda
individual edition, which is distributed free of charge. There is also
the option to upgrade to even more feature-rich team and
enterprise editions for advanced enterprise data analytics
requirements.

Python notebooks
Jupyter Notebooks, have become the defector standard when it
comes to creating notebooks. Notebooks offer users a browser-
based coding environment that can be used to create and share
notebooks that contain everything you need in one spot:

 Code
 Visualizations
 Equations
 Text

Notebooks offer a simpler development experience and a perfect

test platform without the need for setting up dedicated
environments. This is highly useful when dealing with visualization
as some will require complex configurations or additional addons to
display the generated charts, plots, etc.

With notebooks, users can analyze, document, and visualize data in

a single canvas without having to depend on multiple tools.
Notebooks are not limited to Python, and they support over 40
programming languages. This extensive support allows notebooks to
easily provide extended functionality using different languages in
conjunction with a Python notebook.

Machine learning algorithms

All major machine learning algorithms—including TensorFlow,
Microsoft Cognitive Toolkit, scikit-learn, and Spark ML—are written in
Python. All these projects are continuously improved by each
organization and the massive academic and hobbyist communities
backing them. Therefore, users can simply utilize these algorithms
and libraries in their projects to provide analytics and even build
neural networks.