Embedded Files
EXPLORATORY DATA ANALYSIS
EXPLORATORY DATA ANALYSIS
OBJECTIVE
OBJECTIVE
The primary objective of this project is to perform Exploratory Data Analysis (EDA) on a merged dataset containing socio-economic indicators such as GDP per capita, Population, and Literacy Rate for various countries. The aim is to clean the dataset, handle missing values, visualize key patterns, and draw meaningful inferences that reflect global disparities in development, education, and income.
Raw Datasets from Kaggle.
PROBLEM STATEMENT
PROBLEM STATEMENT
The dataset contains multiple real-world inconsistencies such as missing values, mixed data types, and unstructured records, which can hinder accurate analysis. The problem is to prepare and explore this combined dataset effectively in order to understand correlations between economic and social indicators, identify regional disparities in development and detect outliers and patterns that can inform socio-economic policy or further study.
APPROACH
APPROACH
The approach adopted in this project begins with importing and preparing the data using Python’s data handling libraries such as pandas and numpy. Missing values and inconsistencies were resolved through imputation and deletion, ensuring the dataset was clean and analysis-ready. Once prepared, various visualization libraries including matplotlib, seaborn, and plotly were used to create heatmaps, scatter plots, choropleth maps, bubble charts, and pair plots. These visual tools allowed for the exploration of bivariate and multivariate relationships among the variables. Through these visualizations, the project was able to explore patterns such as the correlation between literacy and GDP, the impact of population size on income, and regional disparities in development indicators.
DATA VISUALIZATIONS
DATA VISUALIZATIONS
HEAT MAP
CORRELATION MATRIX OF KEY INDICATORS
The stronger the correlation between the two variables, the larger the number, darker the hue.
The closer the correlation is to 1, the more positively associated they are.
Similarly, with a correlation that is closer to -1, one will fall as the other increases.
For Eg.: Here, a strong positive correlation between GDP per capita and literacy rate implies that higher income may support better education.
A negative correlation between population and GDP per capita may suggest that countries with very large populations tend to have lower GDP per capita.
CHOROPLETH
2022 GDP PER CAPITA BY COUNTRY
This choropleth map gives us a clear overview of the GDP per Capita in each country based on the dataset.
The variation in colors is based on how much the GDP per Capita of the country is.
Countries in Western Europe, North America, and some parts of Asia (e.g., Japan, South Korea) have high GDP per capita.
Sub-Saharan Africa and South Asia generally exhibit lower GDP per capita.
Strong geographic economic disparity is evident.
PAIR PLOT
This is a pairplot.
It allows us to plot pairwise relationships between numerical variables in a dataset.
It helps us understand the data by summarising a large amount of data in a single figure.
SCATTER PLOT
This is a scatter plot
It shows us the outliers which indicate that some countries high GDP & population, very high GDP & low population and some have very high population and moderate GDP.
BUBBLE GRAPH
This is a bubble graph.
The colour of the circle tells us the region(all regions were indexed in the dataset) of the country.
The size of the circle tells us the Population of that particular country.
X axis represents the GDP per capita and Y axis, shows the row index according to the dataset.
BAR GRAPHS
In the graph it can be seen that higher the literacy rate lower is the birth rate and hence we can say that in countries with high literacy rate measures are taken to keep population in control.
The orange colour shows birth rate and blue colour shows literacy rate
This graph represents only for the top 20 countries with highest population.
We see the countries with highest population have higher birthrate than the death rate and hence we see they are one of the most populated countries.
The orange colour shows death rate and blue colour shows birth rate
This graph represents only for the top 20 countries with highest population.
CONCLUSION
CONCLUSION
This project effectively demonstrates how exploratory data analysis can transform raw, inconsistent data into meaningful insights about global socio-economic patterns. The visualizations and statistical relationships revealed that higher literacy rates are generally associated with higher GDP per capita, while population size does not necessarily translate into economic wealth. It also highlighted how countries within the same region can differ significantly in their development metrics. Outliers such as resource-rich but less literate nations, or highly educated yet small economies, further illustrated the complexity of global development. Overall, the project underscores the value of EDA in understanding real-world data and forming the foundation for deeper economic analysis or policy planning.
A PROJECT BY :
KRITIKA NAIK
kritika.naik@eco.christuniversity.in
Page updated
Report abuse