Python for data analysis (ITP and NPV)

Team 2 : MTech Project Presentation (Chethan, Chandra, Divyamsh and Nagabhushan)
Python for Data Analysis
(ITP & NPV Group Project)
Chethan, Chandra, Nagabhushan and Divyamsh
Project Team 2

MTechProjectPresentation
2
Overview of Python Libraries for Data Analysis (Numpy
Pandas Visualisation) part of the presentation)
Data Analysis on ECOMMERCE PURCHASE
Various Data plots using Matplotlib and Seaborn
libraries and other public utilities
Using public libraries like GMPLOT
Introduction to print() Function in Python

Team 2 : MTech Project Presentation (Chethan, Chandra, Divyamsh and Nagabhushan) 3
Summary from the “Ecommerce Purchase” Dataset
The data is predominently a list of purchases from the E-commerse site by a
corporate users
The analysis shows, who bought the most
What professions use this site and to purchase what
What kind of browsers and what kind of Machines they use.
We have also tried to analyse, if the cost of servicing is more for certain states
(least user base)
All the analysis is shown using the graphs.

Which card is being used the most by the customers
Based on this data, the merchant
can decide and promote certain user
base based on the card types,
like the example of offer on HDFC
cards in India (Here they can offer
discounts or tie up with the certain
card companies to increase the
base.
df['CC Provider'].value_counts().plot(kind='pie',autopct='%1.1f%%')
plt.show()
df['CC Provider'].value_counts().plot(kind='bar')
plt.show()
Code snippet

Who are the frequent visitors and who logs in least
Designers, Lawyers are the top 2
people, who visit the site for purchase
Who visit the least
highOcc=df.groupby('Job').sum()['Purchase
Price'].sort_values(ascending=False).head(10)
Code snippet

75% of the cards in the commerce site have expired
chart=sns.countplot(df['CC Exp Date']<"09/20")
Code snippet
The payments/repeat orders on the cards
stored in the
E-commerce may not work as the cards
have expired.
This clearly shows that the frequency of
the users is less
This can also lead to the reduced sales in
the upcoming period

The site should serve the top buying companies
Top 5 companies in terms of purchase
Brown Ltd 15
Smith Group 15
Smith PLC 13
Smith LLC 13
Willia ms LLC 12
Brown Limited and
its spread on cards
the employees use.
Smith Group and its
spread on cards the
employees use.
Code snippet

Technological setup of the users (Promotions strategy)
Code snippet : l1 = df['Browser Info'].str.count('Linux').value_counts()[1]
w1 = df['Browser Info'].str.count('Windows').value_counts()[1]
m1 = df['Browser Info'].str.count('Mac').value_counts()[1]
x = ('Linux','Windows','Mac')
y = (l1,w1,m1)
plt.bar(x,y)
df.loc[df['Browser Info'].str.split('/').str.len() > 1, 'Browser Type'] = df['Browser Info'].str.split('/').str[0]
plt.figure(figsize=(10,6))
Code snippet :

Google Map interface to show the least purchases
The code requires GOOGLE MAP API key, We are directly importing the ﬁle
genrated during the development as this involved cost
Code snippet :
import googlemaps, gmplot, webbrowser, os, json
apikey = '' # Add your google map API key
gmaps = googlemaps.Client(key=apikey) # Instantiating a google map
gmap = gmplot.GoogleMapPlotter.from_geocode('SC, US' , apikey=apikey)
print(type(Lowest_Buying_Cities))
for i in Lowest_Buying_Cities.index:
geocode_result = gmaps.geocode(i)
geom = geocode_result[0]['geometry']
loc = geom['location']
lat = loc['lat']
lng = loc['lng']
hidden_gem_lat, hidden_gem_lon = lat,lng
gmap.marker(hidden_gem_lat, hidden_gem_lon, 'red',title=i,label=i)
# Draw
gmap.draw("Lowest_Buying_Cities.html")

PRINT() Function in Python - Team 2
print(), when called simply throws the cursor to the next line means, a blank line will be displayed
Print(“string”)
Example Output
print() Prints the ‘n’ character
print(“Hello”) Hello
print(‘Hello’) Hello
print(“Hello n World”) Hello
World
print(“Hello t World Hello World
print(“Hello n World”) Hello n World
print(3* ’Hello’) HelloHelloHello
print(“Hello”+”World”) HelloWorld
print(“Hello”,”World”) Hello World
Print(variable list)
Example Output
a, b = 1, 2
print (a,b)
1 2
print(a,b, sep=“,”) 1,2
print(a,b, sep=‘:’) 1:2
print(a,b, sep=‘- - - ‘ 1 - - - 2
print(“Hello”, end=“ “)
print(“World”) HelloWorld
print(“Hello”, end=“t”)
print(“World”)
Hello World

Print(object)
Example Output
list=[10,’A’,”Hai”]
print(list)
[10, ‘A’, ‘Hai’]
d = {10:”Ram”, 20:”Amar”}
print(d)
{10:”Ram”, 20:”Amar”}
PRINT() Function in Python - Team 2 continued
Print(“String”, variable list)
Example Output
a=2
print(a,”: Even Number”)
print(“You typed”, a,”as input”)
2 : Even Number
You typed 2 as input

Print(formatted string) Syntax: print(“formatted string” % (variable list))
Example Output
a = 10
print("The value of a: %i" % a)
The value of a: 10
a, b = 10, 20
print("a: %dtb: %d" % (a, b))
a: 10 b: 20
name = "Ram"
print("Hai %s" % name)
print("Hai (%20s)" % name)
print("Hai (%-20s)" % name)
Hai Ram
Hai ( Ram)
Hai (Ram )
print("%c" % name[2]) m
print("%s" % name[0:2]) Ra
num = 123.345727
print("Num: %f" % num)
print("Num: %8.2f" % num)
Num: 123.345727
Num: 123.35

Print(formatted string)
Example output
a, b, c = 1, 2, 3
print("First= {0}". format(a))
print("First= {0}, Second= {1}". format(a, b))
print("First= {one}, Second= {two}". format(one=a, two=b))
print("First= {}, Second= {}". format(a, b))
First= 1
First= 1, Second= 2
First= 1, Second= 2
First= 1, Second= 2
name, salary = "Ram", 123.45
print("Hello {0}, your salary: {1}". format(name, salary))
print("Hello {n}, your salary: {s}". format(n=name, s=salary))
print("Hello {:s}, your salary: {:.2f}". format(name, salary))
print("Hello %s, your salary: %.2f" % (name, salary))
Hello Ram, your salary: 123.45
Syntax: print("formatted string" % (varaible list))

APPENDIX

Download notebook
15
Available on request in Google Collab

Python Libraries for Data Science
Many popular Python toolboxes/libraries:
• NumPy
• Pandas
Visualization libraries
• matplotlib
• Seaborn
and many more …
16
All these libraries are
installed on the SCC
#Import Python Libraries
import numpy as np
import pandas as pd
import matplotlib as mpl
import seaborn as sns

NumPy:
▪ introduces objects for multidimensional arrays and matrices, as well as
functions that allow to easily perform advanced mathematical and statistical
operations on those objects
▪ provides vectorization of mathematical operations on arrays and matrices
which significantly improves the performance
▪ many other python libraries are built on NumPy
17

Pandas:
▪ adds data structures and tools designed to work with table-like data (similar to
Series and Data Frames in R)
▪ provides tools for data manipulation: reshaping, merging, sorting, slicing,
aggregation etc.
▪ allows handling missing data
18
Link: http://pandas.pydata.org/

matplotlib:
▪ python 2D plotting library which produces publication quality figures in a
variety of hardcopy formats
▪ a set of functionalities similar to those of MATLAB
▪ line plots, scatter plots, barcharts, histograms, pie charts etc.
▪ relatively low-level; some effort needed to create advanced visualization
Link: https://matplotlib.org/
19

Seaborn:
▪ based on matplotlib
▪ provides high level interface for drawing attractive statistical graphics
▪ Similar (in style) to the popular ggplot2 library in R
Link: https://seaborn.pydata.org/
20

Python for data analysis (ITP and NPV)

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Python for data analysis (ITP and NPV)

Similar to Python for data analysis (ITP and NPV) (20)

Recently uploaded

Recently uploaded (20)

Python for data analysis (ITP and NPV)