Forem: HannahMwende

Indian Start-up Funding Analysis

HannahMwende — Sun, 23 Jul 2023 20:25:46 +0000

1. Introduction/Objective 

Funding plays a pivotal role in providing the necessary resources and support for startups to thrive. The project focuses on analyzing the funding received by start-ups in India from 2018 to 2021. The main objective is to gain insights into the Indian start-up ecosystem and propose the best course of action for our team's venture. By analyzing the data on funding amounts, start-up details, and investor information, we aim to unearth prevailing patterns and gain insights about the opportunities in India's start-up ecosystem to inform decision-making. To help deduce various insights the following guiding questions were used: 

1.1 Research Questions

i. What is the overall trend in funding received by start-ups in India from 2018 to 2021? 
ii. Which industries or sectors have received the highest funding during this period? 
iii. What is the distribution of startups across the cities in India? 
iv. What is the average funding amount received by start-ups in India during this period? 
v. Which Investors Funded Most Start-Ups Over the Period?
vi. Which Investors Funded the Most Money on Average Over the Period?
vii. At What Stages were Most Start-ups Funded?
viii. Which 5 Cities Recorded Most Funding?

Project Structure 

The Cross Industry Standard Process for Data Mining (CRISP-DM) framework was fully adopted in conducting the data analysis process on this project. The sequential steps followed were: 

Business understanding. This stage involved setting the project objective. 
Data understanding. This stage verifying the quality of our data (is it clean?) and later exploring and describing our data. 
Data preparation. This stage entailed cleaning, integrating and reformatting our data. 
Data Visualization and storytelling. 
Deployment.

2. Data understanding 

The data used in this project was a combination of 4 different datasets. The dataset encompasses information on the Indian startup ecosystem for the years 2018, 2019, 2020, and 2021. Each dataset contains crucial details about funding stages, companies, founding years, funding amounts, investors, sectors, company information, founders, and headquarters locations. 

3. Data preparation 

During this stage, the datasets were cleaned separately, merged thereafter to form an aggregate dataset used in the analysis. The data preparation stage entailed: 

Detecting and dealing with missing values and duplicates. 
Splitting Location and Industry Columns: The location and industry columns contained multiple values separated by commas. Only the first value was selected as the primary sector.  
Currency Conversion for 2018 Amounts: In the 2018 dataset, the amounts column contained a mix of Indian Rupees (INR) and US Dollars (USD). To standardize the amounts, the Indian Rupee was converted to US Dollars (USD). 
Removing commas and currency signs from the Amount column in all datasets. This allowed the amounts to be properly recognized as numeric values. 
Correcting misplaced/erroneous values especially in the 2021 dataset. 
Dropping an extra column named “column10” in the 2020 dataset: This was to align the columns as this column was not present in other datasets. 
Adding “Funding Year” column to every dataset which would enable analyzing the funding trends over time. 
Renaming Columns and Concatenation. 
Data type conversions (e.g., numeric data mistakenly encoded as strings)

4. Data Visualization and storytelling.

This involves using compelling visuals and charts to present the data in an understandable and meaningful way. To establish a story told by the data on the overall funding of various startups in India, visualizations of various factors that affect the funding were displayed in a dashboard using PowerBI and later published to Power BI service. To help visualize and draw insights the research questions were used as a guide:

Answering the Research Questions

4.1 What is the overall trend in funding received by startups in India from 2018-2021? 

A line graph of Total Funding Amount against the year of funding was plotted. To address this question, we used the ‘groupby’ function in python to group the data according to ‘Year Funded’ and aggregate ‘Amount’ by sum. 

Insights Drawn
For the period 2018 -2019, there was a downward trend in the amount of money funded into startups. From 2019, the money funded has steadily increased. 

4.2 Which industries or sectors have received the highest funding during this period? 

A bar plot of Sector against the total funding amount was plotted. The data was grouped by the Sector column sum of Funding Amount calculated. 

Insights Drawn
From the graph above, Financial Services is the highest funded sector, which is then followed by Commerce and Shopping while the Gaming sector was the least funded. 

4.3 What is the Distribution of Start-ups Across the Cities in India? 

This question answered the total amount of startups across the cities funded across the time period 2018-2021. A bar plot of Cities against the start-up count per city was plotted.

Insights Drawn 
Most funded startups were in Bangalore. This can be explained by most startups having their headquarters located in Bengaluru (Bangalore), which is in the Indian State of Karnataka followed by Mumbai then Gurugram and New Delhi. 

4.5 Which Investors Funded Most Start-Ups Over the Period? 

A bar plot of investor against the startup count was plotted. The data was grouped by the Investor and a total count of startup calculated. 

Insight Drawn 
For most startups, their investors were either Unknown or Undisclosed. We assume it makes sense to expurgate such information from our analysis since it will not be of much benefit to our stakeholders.  

The leading investor that had invested in most startups was therefore Venture Catalysts, which was closely followed by Inflection Point Ventures. Thereafter came Titan Capital, Tiger Global and Mumbai Angels Network. 

4.6 Which Investors Funded the Most Money on Average Over the Period? 

A bar plot of investor against the total amount funded was plotted. The data was grouped by the Investor and sum of funds invested calculated.  

Insight Drawn
Silver Lake and Mubadala Investment Company are the leading investors who funded the highest average amount into startups. 

4.7 At What Stages were Most Start-ups Funded? 

A doughnut chart displaying the top three stages when the companies were funded is plotted. The data was grouped by the Stage and a total count of startup calculated. 

Insight Drawn
Debt financing recorded the highest funding in the Indian start-up ecosystem with a whopping 84.2 percent share among three top stages of funding over the period of 2018 to 2021. This was followed by venture equity whose series was not known then Series C. These were just 8.3% and 7.5% respectively. 

4.8 Which 5 Cities Recorded Most Funding? 

A bar plot of Cities against the total amount funded was plotted. The data was grouped by the Cities and a sum of funds invested calculated. 

Insight Drawn
According to the analysis, the most highly funded start-ups are in Mumbai as it received over 233 billion dollars over the period. The next was Bangalore, which had a little over 24 billion dollars. New Delhi then came with about 5 billion dollars with just minute gaps from Gurugram and those located outside India.

Conclusion

By rigorously examining and exploring the data and conducting these analyses, it was discovered that funding dropped from 2018 to 2019 by over 17 billion dollars before rising steadily to 2021 therefore fail to reject the null hypothesis which states that the funding received by start-ups in India has not demonstrated a consistent upward trajectory over the years.
It also found out that there are significant disparities in funding received during various stages. The debt financing stage received the highest funding over the period with 150.7 billion. Venture capital with unknown series, however, came second with 14.9 billion and Series C with 13.4 billion.

Recommendations

Situation analysis for a venture into the Indian ecosystem regarding funding of start-ups is particularly important as it is highly likely to receive funding as a start-up when headquartered at Mumbai, Bangalore, New Delhi and Gurugram respectively.
Alternative financing sources must be implored to support a start-up since most of the start-ups failed to receive highly significant funding at their initial stages. The earliest highly funded was at the Series C stage when it is expected to be a market fit.
Considering the sector to venture into as far as funding for a start-up in India is concerned is vital as it is more likely to receive funding when at the financial services, commerce and shopping, administrative and education sectors, respectively.

The entire project can be accessed on Github.

Object oriented programming using python

HannahMwende — Wed, 06 Apr 2022 09:28:13 +0000

Python is a great programming language that supports Object Oriented Programming.

What is Object oriented programming?

This is a flexible, powerful paradigm where classes represent and define concepts while objects are instances of classes.

Why use Object Oriented Programming?

It makes code more reusable so you don't have to repeat once
defined.
Makes it easier to work with larger programs.
Makes program easy to understand.

A class is a user-defined blueprint from which objects are created.
Classes have attributes and methods associated with them and you can have instances of a class when you create new, individual objects in a certain class.

For example in the above diagram, we have a class that we have named Fruit.

Banana, apple, pineapple and strawberry are instances of the class Fruit.
An instance is an object that is built from a class and contains real data.

Attributes are the characteristics associated to a class or a certain instance of a class. Like in our example apple can have attributes: color(red) and texture(smooth)

Methods are functions that operate on the attributes of a specific instance of a class. In our apple example ,we can have a cut method which turns it to pieces or eat method which reduces the apple size with every bite.

Creating and defining classes in Python

We create and define classes in Python similar to how we define functions.
Start with the class keyword, followed by the name of the class and a colon. Class names start with a capital letter. After the class definition line is the class body, indented to the right.

class Fruit:
  color=" "
  texture=" "

We can set the attributes of our class instance by accessing them using dot notation.
Dot notation can be used to set or retrieve object attributes as well as call methods associated with them.

apple.color="red"
apple.texture="smooth"

We created a Fruit instance called apple and set the color and texture attributes. Another instance of Fruit can be created and set different attributes.

pineapple.color="brown"
pineapple.texture="rough"

Special methods

The examples above are classes and objects in their simplest form and are not really useful in real life applications.

Special methods start and end with two underscore characters.
There are many special methods that you can use to customize classes in Python.

All classes have a special method __ init __() ,which is always executed when the class is being used to create a new object. It is used to assign values to object properties or other operations that are necessary to do when the object is being created.

class Fruit:
  def __init__(self,name,color,texture):
    self.name=name
    self.color=color
    self.texture=texture

fruit1=Fruit("apple","red","smooth")

print(fruit1.color)
print(fruit1.texture)
print(fruit1.name)

Output:

red
smooth
apple

The self parameter is a reference to the current instance of
the class, and is used to access variables that belong to the
class.

There is also a __ str __() special method which returns the output in a string format.

class Fruit:
  def __init__(self,name,color,texture):
    self.name=name
    self.color=color
    self.texture=texture



  def __str__(self):
    return f"{self.name} is a fruit and its {self.color} in color and its {self.texture}"

fruit2=Fruit("Apple","red","smooth")
print(fruit2)

Output:

Apple is a fruit and its red in color and its smooth

An example of method in our Fruit class is as shown below:

class Fruit:
  def __init__(self,name,color,texture):
    self.name=name
    self.color=color
    self.texture=texture

  def fruity(self):
    return f"{self.name} is {self.color} in color and {self.texture} "

  def describe(self,taste):
    return f"{self.name} is {taste}"


fruit1=Fruit("apple","red","smooth")
print(fruit1.fruity())

print(fruit1.describe("sweet"))

Output:

apple is red in color and smooth 
apple is sweet

Fruity and describe are methods in our Fruit class.

There are several key concepts in Object Oriented Programming.

Inheritance

In object-oriented programming, the concept of inheritance allows you to build relationships between objects, group together similar concepts and reduce code duplication.

A parent class is the class being inherited from.Any class can be a parent class, so the syntax is the same as creating any other class.

A child class is the class that inherits from another class.A child class inherits the properties and methods from its parent but you can also add other properties.
When creating a child class, the name of the parent class is set as the parameter so that it can inherit its functionality.

To keep the inheritance of the parent's init() function,
add a call to the parent's init() function

class Person:
  def __init__(self,name,age):
    self.name = name
    self.age = age

  def display(self):
    print(self.name+" is "+str(self.age)+" years old")

#child class
class Student(Person):
  def __init__(self,name,age,admission,form):
    self.admission=admission
    self.form = form

    # invoking the __init__ of the parent class 
    Person.__init__(self,name,age)


#Creating an object
Aziz = Student("Abdul Aziz","22","6325",4)

Aziz.display()

Output
>>> Abdul Aziz is 22 years old

From the above example, we have created a child class Student from the parent class Person and it inherits the attributes name and age and method display from its parent.Although, we have added other attributes admission and form to our child class.

Python also has a super() function that will make the child class inherit all the methods and properties from its parent

Encapsulation

This concept allows us to restrict access to methods and variables. This prevents data from direct modification. In Python, we denote private attributes using underscore which can either be single _ or double __.

class Computer:

    def __init__(self):
        self.__maxprice = 900

    def sell(self):
        print("Selling Price: {}".format(self.__maxprice))

    def setMaxPrice(self, price):
        self.__maxprice = price

c = Computer()
c.sell()

# change the price
c.__maxprice = 1000
c.sell()

# using setter function
c.setMaxPrice(1000)
c.sell()

Output:

Selling Price: 900
Selling Price: 900
Selling Price: 1000

In the above program ,we have defined a computer class whereby the_init_() method is for storing the maximum selling price of computer.
We try to set max price outside and it cannot be reflected since its a private variable __maxprice hence using a setter function to change the price.

Polymorphism

The word polymorphism means having many forms. In programming, polymorphism means the same function name (but performing different activities) which allows flexibility.

class India():
    def capital(self):
        print("New Delhi is the capital of India.")

    def language(self):
        print("Hindi is the most widely spoken language of India.")

    def type(self):
        print("India is a developing country.")

class USA():
    def capital(self):
        print("Washington, D.C. is the capital of USA.")

    def language(self):
        print("English is the primary language of USA.")

    def type(self):
        print("USA is a developed country.")

obj_ind = India()
obj_usa = USA()
for country in (obj_ind, obj_usa):
    country.capital()
    country.language()
    country.type()

Output:

New Delhi is the capital of India.
Hindi is the most widely spoken language of India.
India is a developing country.
Washington, D.C. is the capital of USA.
English is the primary language of USA.
USA is a developed country.

We have created two classes India and USA .They have similar structure and methods capital,language,type although they are not linked in any way, we are able to use a for loop to iterate them through a common country variable.
This is made possible by polymorphism.

For more understanding of polymorphism concept here are some links:
https://www.geeksforgeeks.org/polymorphism-in-python/
https://www.programiz.com/python-programming/polymorphism

Introduction to Data Structures and Algorithms With Python

HannahMwende — Mon, 21 Feb 2022 07:12:45 +0000

Data Structures are a collection of data values, the relationships among them, and the functions or operations that can be applied to the data.

Data Structures allow you to organize your data in such a way that enables you to store collections of data, relate them and perform operations on them accordingly.

Data algorithms are sets of well-defined instructions to solve a particular problem. It takes a set of input and produces a desired output.

Types of data structures

1. Lists

This is an ordered and changeable collection of item that also allows duplicate items. It can contain different data types.
Lists are created through square brackets[] or list() constructor.
For example:

my_list=["melon","apple","mango"]

List Methods
Python has a set of built-in methods that you can use on lists.

#access a list item
print(my_list[2])

#append()  Adds an element at the end of the list
my_list.append("orange")

#clear()  Removes all the elements from the list
my_list.clear()

#copy()  Returns a copy of the list
my_list.copy()

#extend()Adds list items to the end of the current list 

my_list.extend(another_list)

#insert()  Adds an element at the specified position\index
my_list.insert(3,"guava")

#pop()  Removes the element at the specified position\index
my_list.pop(1)

#remove()  Removes the item with the specified value
my_list.remove("apple")

#reverse()  Reverses the order of the list
my_list.reverse()

#sort()  Sorts the list alphanumerically, ascending, by default
my_list.sort()

2. Tuples

These are unchangeable,ordered data structures that allows duplicates.
They are created through () or tuple() constructor.

#create a tuple
number=(45,34,678)

#access a tuple item
number[0]

#delete a tuple
del number

Once you have created a tuple you can not change it, that is you can not add, edit or remove any of the items in it. Not unless you change it to a mutable data structure like lists then back to a tuple.

y=list(number)
#add the list method you want performed e.g sort()
y.sort()
number=tuple(y)

3. Dictionary

Dictionaries are used to store data values in key:value pairs.
They are changeable and do not allow duplicate keys and are written with curly brackets {}

dog = {'name':'Cara',
'color':'Brown',
'gender':'Female'}

Dictionary methods

#keys()  prints out all the keys
for i in dog.keys():
   print(i)

#values()  prints out all the values
for i in dog.values():
   print(i)

#items() prints out the key:value pair
for i in dog.items():
   print(i)

Output

name
color
gender

Cara
Brown
Female

('name', 'Cara')
('color', 'Brown')
('gender', 'Female')

It’s tedious to check whether a key exists in a dictionary before accessing that key’s value. Fortunately, dictionaries have a get() method that takes two arguments: the key of the value to retrieve and a fallback value to return if that key does not exist.

print(dog.get('sound',0))

Output

4. Sets

Unlike dictionaries,sets are unchangeable and also do not support duplicates.

myset = {"apple", "banana", "cherry"}

5. Stacks

A stack is a linear data structure that follows the principle of Last In First Out (LIFO). This means the last element inserted inside the stack is removed first.
In programming terms, putting an item on top of the stack is called push and removing an item is called pop.

# Creating a stack
def create_stack():
    stack = []
    return stack


# Creating an empty stack
def check_empty(stack):
    return len(stack) == 0


# Adding items into the stack
def push(stack, item):
    stack.append(item)
    print("pushed item: " + item)


# Removing an element from the stack
def pop(stack):
    if (check_empty(stack)):
        return "stack is empty"

    return stack.pop()


stack = create_stack()
push(stack, str(1))
push(stack, str(2))
push(stack, str(3))
push(stack, str(4))
print("popped item: " + pop(stack))
print("stack after popping an element: " + str(stack))

Output

pushed item: 1
pushed item: 2
pushed item: 3
pushed item: 4
popped item: 4
stack after popping an element: ['1', '2', '3']

6. Queues

A queue is a linear type of data structure used to store the data in a sequentially and it follows the First In First Out (FIFO) rule - the item that goes in first is the item that comes out first.
Enqueue is adding an element to the end of the queue while
Dequeue is removing an element from the front of the queue.

# Queue implementation in Python

class Queue:

    def __init__(self):
        self.queue = []

    # Add an element
    def enqueue(self, item):
        self.queue.append(item)

    # Remove an element
    def dequeue(self):
        if len(self.queue) < 1:
            return None
        return self.queue.pop(0)

    # Display  the queue
    def display(self):
        print(self.queue)

    def size(self):
        return len(self.queue)


q = Queue()
q.enqueue(1)
q.enqueue(2)
q.enqueue(3)
q.enqueue(4)
q.enqueue(5)

q.display()

q.dequeue()

print("After removing an element")
q.display()

Output

[1, 2, 3, 4, 5]
After removing an element
[2, 3, 4, 5]

7. Linked lists

A linked list is a linear data structure that includes a series of connected nodes. Each node stores the data and the address of the next node.
The first node is called the head, and it’s used as the starting point for any iteration through the list. The last node must have its next reference pointing to None to determine the end of the list. Here’s how it looks:

# Linked list implementation in Python


class Node:
    # Creating a node
    def __init__(self, item):
        self.item = item
        self.next = None


class LinkedList:

    def __init__(self):
        self.head = None


if __name__ == '__main__':

    linked_list = LinkedList()

    # Assign item values
    linked_list.head = Node(1)
    second = Node(2)
    third = Node(3)

    # Connect nodes
    linked_list.head.next = second
    second.next = third

    # Print the linked list item
    while linked_list.head != None:
        print(linked_list.head.item, end=" ")
        linked_list.head = linked_list.head.next

Output

1 2 3

So what makes them different from normal lists? Linked lists differ from lists in the way that they store elements in memory. While lists use a contiguous memory block to store references to their data, linked lists store references as part of their own elements.

The above concepts can be overwhelming ,so just take your time.
Happy coding!!

Introduction to modern python

HannahMwende — Sun, 13 Feb 2022 16:19:14 +0000

Python is an interpreted, high-level programming language with dynamic semantics. Python programs have the extension.py and can be run from the command line by typing python file_name.py.

Why learn Python?

It is also quite easy to learn, understand, code and a developer-friendly language.
Python is free and open sourced. Not only can you download Python for free, but you can also download its source code, make changes to it and even distribute it.
Python supports object-oriented programming and concepts of classes, objects encapsulation, etc.
Python can be extended to other languages. You can write some of your code in languages like C++ or C. This comes in handy, especially in projects.
Easily portable as it can be shared and you can run a code on any platform.
Python has extensive libraries that contain codes for various purposes.So, we don’t have to write the complete code for that manually.

Where can Python be used ?

Artificial Intelligence and machine learning.
Web development.
Game development.
Data science.
Audio and Visual Application.
Software Development.
Automation.

Setup your python working environment

For starters or data science enthusiasts,install anaconda

Anaconda | Individual Edition
Anaconda's open-source Individual Edition is the easiest way to perform Python/R data science and machine learning on a single machine. (19 kB)

For software developers/engineers install python interpreter ( preferably Python 3.8) here

Also install a text editor of your choice.I prefer Visual Studio Code which can be downloaded through this link.

At the end of the installation,confirm your installation by opening the terminal and type:

python --version

Then it should print the python version you just installed.
Now we are all set and you can proceed to create your first program!

Launch VS Code and create a new file where your codes will be saved and it must have the .py extension.
For instance,I will create a file called helloworld.py and type in :

print("Hello World!")

Save your file. Open your command line, navigate to the directory where you saved your file, and run the code.
The output should be:

Hello World!

Congratulations, you have written and executed your first Python program.

Forem: HannahMwende

Indian Start-up Funding Analysis

1. Introduction/Objective

1.1 Research Questions

Project Structure

2. Data understanding

3. Data preparation

4. Data Visualization and storytelling.

4.1 What is the overall trend in funding received by startups in India from 2018-2021?

4.2 Which industries or sectors have received the highest funding during this period?

4.3 What is the Distribution of Start-ups Across the Cities in India?

4.5 Which Investors Funded Most Start-Ups Over the Period?

4.6 Which Investors Funded the Most Money on Average Over the Period?

4.7 At What Stages were Most Start-ups Funded?

4.8 Which 5 Cities Recorded Most Funding?

Conclusion

Recommendations

Object oriented programming using python

What is Object oriented programming?

Why use Object Oriented Programming?

Creating and defining classes in Python

Special methods

Inheritance

Encapsulation

Polymorphism

Introduction to Data Structures and Algorithms With Python

Types of data structures

1. Lists

2. Tuples

3. Dictionary

4. Sets

5. Stacks

6. Queues

7. Linked lists

Introduction to modern python

Why learn Python?

Where can Python be used ?

Setup your python working environment

1. Introduction/Objective 

Project Structure 

2. Data understanding 

3. Data preparation 

4.1 What is the overall trend in funding received by startups in India from 2018-2021? 

4.2 Which industries or sectors have received the highest funding during this period? 

4.3 What is the Distribution of Start-ups Across the Cities in India? 

4.5 Which Investors Funded Most Start-Ups Over the Period? 

4.6 Which Investors Funded the Most Money on Average Over the Period? 

4.7 At What Stages were Most Start-ups Funded? 

4.8 Which 5 Cities Recorded Most Funding? 