To perform basic analysis using Python, you'll primarily use libraries like
Pandas, NumPy, and Matplotlib or Seaborn for data handling, manipulation,
and visualization.
Here's a simple guide to get you started.
1. Install Required Libraries
If you don't already have the libraries installed, you can install them using
pip:
code
pip install pandas numpy matplotlib seaborn
2. Loading Data
First, import the necessary libraries and load the data. You can load data
from various formats like CSV, Excel, etc.
Example for loading a CSV file:
python
import pandas as pd
Load dataset
df = pd.read_csv('your_data.csv')
3. Explore the Data
You can perform some basic exploration to understand the data.
- Check the first few rows of the dataset:
python
df.head()
```
- Get basic info about data types and missing values:
python
df.info()
- Get summary statistics:
python
df.describe()
4. Data Cleaning
This step often involves handling missing data, duplicates, or fixing data
types.
- Handle missing data by filling or dropping:
python
df.fillna(0, inplace=True) # Fill missing values with 0
df.dropna(inplace=True) # Drop rows with missing values
- Drop duplicates :
python
df.drop_duplicates(inplace=True)
5. Basic Analysis
You can begin with basic descriptive statistics and visualizations.
a. Descriptive Statistics
- Mean, median, mode:
python
mean_value = df['column_name'].mean()
median_value = df['column_name'].median()
mode_value = df['column_name'].mode()[0]
- Value counts (for categorical variables):
python
df['category_column'].value_counts()
b. Group By Analysis
You can group data by a particular column and perform aggregate
operations.
python
grouped_data = df.groupby('category_column')['numerical_column'].sum()
c. Correlation
Check the correlation between numerical features.
python
correlation_matrix = df.corr()
print(correlation_matrix)
6. Basic Data Visualization
Visualization is key to data analysis.
a. Histograms
To visualize the distribution of a column:
python
import matplotlib.pyplot as plt
df['column_name'].hist()
plt.show()b. Scatter Plot
To check the relationship between two variables:
python
df.plot(kind='scatter', x='column1', y='column2')
plt.show()
c. Box Plot
To identify outliers:
python
df.boxplot(column='numerical_column')
plt.show()
d. Correlation Heatmap (using Seaborn)
For a more visual representation of correlation:
python
import seaborn as sns
sns.heatmap(df.corr(), annot=True, cmap='coolwarm')
plt.show()
7. Saving Cleaned Data
After cleaning and analysis, you might want to save the processed data.
python
df.to_csv('cleaned_data.csv', index=False)
Example Workflowpython
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
Load data
df = pd.read_csv('your_data.csv')
Basic exploration
print(df.head())
print(df.info())
print(df.describe())
Handle missing values
df.fillna(0, inplace=True)
Descriptive statistics
print(df['age'].mean()) # Example for 'age' column
print(df['category'].value_counts()) # For categorical data
Visualize data
df['age'].hist()
plt.show()
sns.heatmap(df.corr(), annot=True, cmap='coolwarm')
plt.show()
This workflow should get you started on basic data analysis using Python!
You can further enhance this by using more advanced libraries like SciPy for
statistical analysis or StatsModels for regression and other statistical
models.
