COVID-19 Dataset Analysis Project

 

COVID-19 Dataset Analysis Project (Using Kaggle Dataset)

(Python · Pandas · Matplotlib)

---

1. Project Title

COVID-19 Data Analysis Using Python (Exploratory Data Analysis on Global COVID Statistics)

---

2. Dataset Source

Download from Kaggle:
“Novel Corona Virus 2019 Dataset” or
“COVID-19 World Vaccination Progress”
Search on Kaggle → Download CSV.

You will mainly use:

• covid_19_data.csv

• time_series_covid_19_confirmed.csv

• time_series_covid_19_deaths.csv

• time_series_covid_19_recovered.csv

---

3. Project Objectives

✔ Understand global COVID-19 spread through data
✔ Identify most affected countries
✔ Visualize daily vs cumulative cases
✔ Analyze death & recovery trends
✔ Explore correlation between features
✔ Perform country-wise time-series analysis

---

4. Python Libraries Required

pip install pandas matplotlib seaborn numpy

---

5. Import Libraries

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

---

6. Load Dataset

df = pd.read_csv("covid_19_data.csv")
df.head()

---

7. Basic Data Information

df.info()
df.isnull().sum()
df.describe()

---

8. Data Cleaning

Rename messy column names:

df.rename(columns={
    'Country/Region': 'Country',
    'Province/State': 'State',
    'ObservationDate': 'Date'
}, inplace=True)

Convert date:

df['Date'] = pd.to_datetime(df['Date'])

Fill missing values:

df['State'] = df['State'].fillna('')

---

9. Global Numbers Overview

Total Confirmed, Deaths, Recovered

total_confirmed = df['Confirmed'].sum()
total_deaths = df['Deaths'].sum()
total_recovered = df['Recovered'].sum()

print(total_confirmed, total_deaths, total_recovered)

---

10. Country-wise Summary

country_summary = df.groupby("Country")[["Confirmed","Deaths","Recovered"]].max().sort_values("Confirmed", ascending=False)
country_summary.head(10)

---

11. Plot: Top 10 Countries by Confirmed Cases

top10 = country_summary.head(10)

plt.figure(figsize=(10,6))
plt.bar(top10.index, top10['Confirmed'])
plt.xticks(rotation=45)
plt.title("Top 10 Countries with Highest Confirmed Cases")
plt.xlabel("Country")
plt.ylabel("Confirmed Cases")
plt.show()

---

12. Death Rate and Recovery Rate

country_summary["Death_Rate"] = (country_summary["Deaths"] / country_summary["Confirmed"]) * 100
country_summary["Recovery_Rate"] = (country_summary["Recovered"] / country_summary["Confirmed"]) * 100

country_summary.head()

---

13. Heatmap (Correlation)

plt.figure(figsize=(6,4))
sns.heatmap(df[['Confirmed','Deaths','Recovered']].corr(), annot=True)
plt.title("Correlation Between COVID-19 Metrics")
plt.show()

---

14. Time Series Analysis (Global Trend)

global_daily = df.groupby("Date")[["Confirmed","Deaths","Recovered"]].sum()

plt.figure(figsize=(12,6))
plt.plot(global_daily.index, global_daily["Confirmed"], label="Confirmed")
plt.plot(global_daily.index, global_daily["Deaths"], label="Deaths")
plt.plot(global_daily.index, global_daily["Recovered"], label="Recovered")
plt.legend()
plt.title("Global Trend of COVID-19 Cases Over Time")
plt.xlabel("Date")
plt.ylabel("Cases")
plt.show()

---

15. Country-Specific Analysis (e.g., India)

india = df[df["Country"]=="India"].groupby("Date")[["Confirmed","Deaths","Recovered"]].sum()

plt.figure(figsize=(10,5))
plt.plot(india.index, india["Confirmed"], label="Confirmed")
plt.plot(india.index, india["Deaths"], label="Deaths")
plt.plot(india.index, india["Recovered"], label="Recovered")
plt.title("COVID-19 Trend in India")
plt.legend()
plt.show()

---

16. Key Findings (Write in Report)

✔ USA, India, Brazil were most affected
✔ Confirmed cases show exponential growth in early months
✔ High correlation between confirmed cases and deaths
✔ Death rate varies per country (health system differences)
✔ Recovery rate increased after vaccination rollout
✔ India shows rapid growth during second wave