Video games sales analysing using in Python using Machine Learning.

By Sudipta Ghosh

The project aim is to analysing top Video Games Sales in Europe, North America, and Japan and also to see top publishers in that countries using Python

Step: 1

First, we will import the necessary libraries then we import the datasets

import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from scipy import stats

Video_Games_sales = pd.read_csv("Video_Games_sales.csv")
Video_Games_sales

Step: 2

Now we will see in which countries the video games sales most

Country_List = ['NA_Sales', 'EU_Sales', 'JP_Sales', 'Other_Sales']
labels = Country_List
plt.figure(figsize=(10,10))
sizes = Video_Games_sales[Country_List].sum()
plt.pie(sizes, labels=labels, autopct='%1.1f%%',shadow=True, startangle=90)
plt.show()

Here we see North America is at the top with 49.3% sales

Step: 3

Now we will see top Games and top publishers in North America

#North America top games
Top = Video_Games_sales.groupby(by=['Name'])['NA_Sales'].sum()
Top = Top.sort_values(ascending = False).nlargest(10)

plt.figure(figsize=(10,10))
labels = Top.index
sizes = Top
explode = (0, 0.1, 0.1, 0.1, 0, 0, 0, 0, 0, 0)
colors = ['darkred', 'maroon', 'firebrick', 'indianred', 'white', 'whitesmoke', 'lightsteelblue', 'cornflowerblue', 'royalblue', 'navy']



plt.pie(sizes, explode = explode, labels=labels, colors = colors, autopct='%1.1f%%',
        shadow=True, startangle=90)

plt.show()

#North America Top publiser
Top = Video_Games_sales.groupby(by=['Publisher'])['NA_Sales'].sum()
Top = Top.sort_values(ascending = False).nlargest(10)

plt.figure(figsize=(8,8))
labels = Top.index
sizes = Top
explode = (0, 0, 0, 0, 0, 0, 0, 0, 0, 0)
colors = ['indianred', 'firebrick', 'maroon', 'darkred', 'white', 'whitesmoke', 'navy', 'royalblue', 'cornflowerblue', 'lightsteelblue']
colors.reverse()

plt.pie(sizes, explode = explode, labels=labels, colors = colors, autopct='%1.1f%%',
        shadow=True, startangle=90)
plt.show()

Step: 4

Now we will see top Games and top publishers in Europe

#Europe top games
Top = Video_Games_sales.groupby(by=['Name'])['EU_Sales'].sum()
Top = top.sort_values(ascending = False).nlargest(10)

plt.figure(figsize=(10,10))
labels = Top.index
sizes = Top
explode = (0, 0, 0, 0.1, 0, 0.1, 0.1, 0, 0, 0)
colors = ['blue', 'mediumblue', 'royalblue','cornflowerblue', 'lightsteelblue','honeydew','lightgoldenrodyellow','lightyellow','yellow','gold']

plt.pie(sizes, explode = explode, labels=labels, colors = colors, autopct='%1.1f%%',
        shadow=True, startangle=90)

plt.show()

#Europe top game publiser
Top = Video_Games_sales.groupby(by=['Publisher'])['EU_Sales'].sum()
Top = Top.sort_values(ascending = False).nlargest(10)

plt.figure(figsize=(10,10))
labels = Top.index
sizes = Top
explode = (0, 0.1, 0, 0, 0, 0, 0, 0, 0, 0)
colors = ['blue', 'mediumblue', 'royalblue','cornflowerblue', 'lightsteelblue','honeydew','lightgoldenrodyellow','lightyellow','yellow','gold']
colors.reverse()
plt.pie(sizes, explode = explode, labels=labels, colors = colors, autopct='%1.1f%%',
        shadow=True, startangle=90)


plt.show()

Step: 5

Now we will see top Games and top publishers in Japan

# Japan top Games
Top = Video_Games_sales.groupby(by=['Name'])['JP_Sales'].sum()
Top = Top.sort_values(ascending = False).nlargest(10)

plt.figure(figsize=(10,10))
labels = Top.index
sizes = Top
explode = (0.1, 0.1, 0, 0, 0.1, 0, 0.1, 0, 0.1, 0)
colors = ['darkred','maroon','firebrick','brown','indianred','lightcoral','peachpuff','seashell','oldlace','whitesmoke']



plt.pie(sizes, explode = explode, labels=labels, colors = colors, autopct='%1.1f%%',
        shadow=True, startangle=90)

plt.show()

#top japan games publiser
Top = Video_Games_sales.groupby(by=['Publisher'])['JP_Sales'].sum()
Top = Top.sort_values(ascending = False).nlargest(10)

plt.figure(figsize=(10,10))
labels = Top.index
sizes = Top
explode = (0.1, 0, 0, 0, 0, 0, 0, 0, 0, 0)
colors = ['darkred','maroon','firebrick','brown','indianred','lightcoral','peachpuff','seashell','oldlace','whitesmoke']



plt.pie(sizes, explode = explode, labels=labels, colors = colors, autopct='%1.1f%%',
        shadow=True, startangle=90)

plt.show()