Data Visualization in R Complete 20 Pages Notes
Uploaded by
Kajal SinghData Visualization in R Complete 20 Pages Notes
Uploaded by
Kajal SinghData Visualization in R Language
Comprehensive Academic Notes (20+ Pages)
1. Introduction to Data Visualization
Data Visualization is the graphical representation of data using visual elements like charts, graphs,
It helps in understanding trends, patterns, correlations, and outliers.
In Data Science, visualization is a core component of Exploratory Data Analysis (EDA).
Objectives:
- Simplify complex datasets
- Identify relationships between variables
- Detect anomalies
- Support decision-making
2. Importance in Business and Research
Visualization helps managers and analysts make data-driven decisions.
Applications include:
- Sales analysis
- Market research
- Financial forecasting
- Customer segmentation
- Risk management
Good visualization improves clarity, reduces misinterpretation, and enhances communication.
3. Graphics Systems in R
R provides multiple graphics systems:
1. Base R Graphics – Built-in plotting functions.
2. ggplot2 – Based on Grammar of Graphics.
3. Lattice – For multivariate data visualization.
4. Plotly – Interactive visualization.
Each system has its own advantages depending on the requirement.
4. Base R Graphics
Basic Plot Syntax:
plot(x, y)
Important parameters:
- main: Title
- xlab, ylab: Axis labels
- col: Color
- pch: Point character
- type: Type of plot (p, l, b, etc.)
Example:
x <- 1:10
y <- x^2
plot(x, y, type="b", col="blue", main="Quadratic Plot")
5. Types of Plots in R
Scatter Plot – Shows relationship between two numeric variables.
Line Plot – Used for time series data.
Bar Plot – Used for categorical data.
Histogram – Frequency distribution of continuous data.
Box Plot – Shows median, quartiles, and outliers.
Pie Chart – Shows proportional distribution.
6. Histogram in Detail
Histogram divides data into bins.
Syntax:
hist(data, breaks=10, col="skyblue")
Key Terms:
- Frequency
- Density
- Bins
- Range
Histograms help understand distribution shape: normal, skewed, uniform.
7. Boxplot in Detail
Boxplot components:
- Minimum
- Q1 (25%)
- Median (50%)
- Q3 (75%)
- Maximum
- Outliers
Syntax:
boxplot(data, col="orange")
8. Introduction to ggplot2
Developed by Hadley Wickham.
Based on Grammar of Graphics.
Install:
[Link]("ggplot2")
library(ggplot2)
Basic Structure:
ggplot(data, aes(x, y)) + geom_point()
9. Grammar of Graphics
Components:
- Data
- Aesthetic mappings (aes)
- Geometric objects (geom_)
- Statistical transformations
- Scales
- Coordinates
- Facets
- Themes
This layered approach makes visualization flexible and powerful.
10. Scatter Plot using ggplot2
Example:
ggplot(mtcars, aes(x=wt, y=mpg)) +
geom_point(color="red") +
ggtitle("Weight vs MPG")
11. Bar Plot using ggplot2
Example:
ggplot(mtcars, aes(x=factor(cyl))) +
geom_bar(fill="purple")
12. Histogram using ggplot2
Example:
ggplot(mtcars, aes(x=mpg)) +
geom_histogram(bins=10, fill="green")
13. Boxplot using ggplot2
Example:
ggplot(mtcars, aes(x=factor(cyl), y=mpg)) +
geom_boxplot(fill="orange")
14. Faceting
Facet allows multiple plots.
facet_wrap(~variable)
facet_grid(row~column)
Useful for comparing categories.
15. Themes and Customization
Themes:
- theme_minimal()
- theme_classic()
- theme_bw()
Customization:
- scale_color_manual()
- labs()
- theme()
16. Interactive Visualization with Plotly
Install:
[Link]("plotly")
library(plotly)
Example:
plot_ly(data=mtcars, x=~wt, y=~mpg, type='scatter', mode='markers')
17. Data Visualization Best Practices
1. Choose appropriate chart.
2. Avoid clutter.
3. Use clear labels.
4. Maintain consistent scales.
5. Highlight key insights.
6. Avoid misleading representations.
18. Time Series Visualization
Used for stock prices, sales over time.
Use type="l" in Base R or geom_line() in ggplot2.
Example:
ggplot(data, aes(x=date, y=sales)) +
geom_line()
19. Correlation and Regression Visualization
Use scatter plot with regression line:
ggplot(data, aes(x, y)) +
geom_point() +
geom_smooth(method="lm")
20. Applications in Data Science
- Business Intelligence
- Healthcare analytics
- Marketing analysis
- Financial modeling
- Machine learning visualization
Visualization is essential before modeling and after prediction for result interpretation.